APPLICATION OF MACHINE VISION
TECHNOLOGY IN DEFECT DETECTION OF
HIGH-PERFORMANCE PHASE NOISE
MEASUREMENT CHIPS
Jing Zhou*
School of Pharmaceutical Business, Zhejiang Pharmaceutical University, Ningbo,
Zhejiang, 315100, China.
zhoujing_calla@163.com
Reception: 15/05/2023 Acceptance: 29/06/2023 Publication: 23/07/2023
Suggested citation:
Zhou, J. (2023). Application of machine vision technology in defect
detection of high-performance phase noise measurement chips. 3C
Tecnología. Glosas de innovación aplicada a la pyme, 12(2), 347-362. https://
doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
347
ABSTRACT
The problem of chip defects has always existed in industrial production, and since
there are more and more environmental problems caused by chip defects, people
have attached greater importance to the identification and detection of chip defects.
Pursuant to the ecological environmental problems caused by chip defects in the
process of chip production, this paper uses machine vision technology to detect the
defects of high-performance phase noise measurement chips. The results suggest
that the accuracy of machine vision technology for the identification of chip defects
reaches up to 98%. The production volume of organic waste gas decreases from
5968.0t/a to 4000t/a. The yield of organic wastewater decreases from 5496m3/d to
4600m3/d. The production amount of solid waste reduces from 8000t/a to 6500t/a.
The aforementioned data all confirm that machine vision technology has the
advantages of automation, high detection efficiency, and high accuracy of defect
identification for the defect detection of high-performance phase noise measurement
chips. And also, by improving the chip defects, the discharge volume of waste gas,
wastewater, and solid waste in the chip production process is reduced, and thereupon
the ecological environment is ameliorated.
KEYWORDS
Machine vision technology; Chips; Defect detection; Environmental pollution;
Ecological environment
INDEX
ABSTRACT
KEYWORDS
1. INTRODUCTION
2. ENVIRONMENTAL POLLUTION CAUSED BY CHIP DEFECTS
2.1. Exhaust gas and solid waste generation due to chip defects
2.2. Chip defect wastewater generation
3. APPLICATION OF MACHINE VISION-BASED TECHNOLOGY FOR HIGH-
PERFORMANCE PHASE NOISE MEASUREMENT CHIP DEFECT DETECTION
4. RESULTS AND ANALYSIS
5. DISCUSSION
6. CONCLUSION
REFERENCES
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
348
ABSTRACT
The problem of chip defects has always existed in industrial production, and since
there are more and more environmental problems caused by chip defects, people
have attached greater importance to the identification and detection of chip defects.
Pursuant to the ecological environmental problems caused by chip defects in the
process of chip production, this paper uses machine vision technology to detect the
defects of high-performance phase noise measurement chips. The results suggest
that the accuracy of machine vision technology for the identification of chip defects
reaches up to 98%. The production volume of organic waste gas decreases from
5968.0t/a to 4000t/a. The yield of organic wastewater decreases from 5496m3/d to
4600m3/d. The production amount of solid waste reduces from 8000t/a to 6500t/a.
The aforementioned data all confirm that machine vision technology has the
advantages of automation, high detection efficiency, and high accuracy of defect
identification for the defect detection of high-performance phase noise measurement
chips. And also, by improving the chip defects, the discharge volume of waste gas,
wastewater, and solid waste in the chip production process is reduced, and thereupon
the ecological environment is ameliorated.
KEYWORDS
Machine vision technology; Chips; Defect detection; Environmental pollution;
Ecological environment
INDEX
ABSTRACT
KEYWORDS
1. INTRODUCTION
2. ENVIRONMENTAL POLLUTION CAUSED BY CHIP DEFECTS
2.1. Exhaust gas and solid waste generation due to chip defects
2.2. Chip defect wastewater generation
3. APPLICATION OF MACHINE VISION-BASED TECHNOLOGY FOR HIGH-
PERFORMANCE PHASE NOISE MEASUREMENT CHIP DEFECT DETECTION
4. RESULTS AND ANALYSIS
5. DISCUSSION
6. CONCLUSION
REFERENCES
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
1. INTRODUCTION
In the process of chip production and manufacturing, the processes are interlinked.
The technology is complex, and slight variations in materials, environment, process
parameters, and other factors often lead to defects in the chip and affect the product
yield. The problem of chip defects has always existed in industrial production [1]. And
there are more and more environmental problems due to chip defects [2]. Therefore,
more and more attention has been paid to the identification and detection of chip
defects. The traditional chip defect identification and detection approach relies on the
manual operation of professional technicians. This method is not only inefficient but
also relies on the subjective judgment of the operator, and the accuracy of detection is
difficult to be guaranteed. The combination of non-destructive testing equipment and
industrial production lines not only ensures the quality of products but also reduces
the cost of manual inspection and improves the efficiency of production. Later, along
with the rapid rise of machine vision technology, many scholars gradually extended
the application of this technology in the field of chip defect detection, making the chip
defect detection method more and more mature and perfect [3].
The traditional chip defect detection technology is usually used to detect and
identify the defect information such as cracks, white spots, defects, and internal
defects contained in the target chip sample by inspection methods such as magnetic
particle inspection method, penetrant inspection method, eddy current inspection
method, ultrasonic inspection method, and X-ray inspection [4-10]. Despite the
achievements of traditional chip defect detection methods, there are some drawbacks.
First, the chip and shape of the sample under inspection are demanding, and second,
the professional requirements for the operator are high. For example, the detection
results of penetrant flaw detection by the operator's influence, X-ray flaw detection if
improper operation will produce radiation hazards to the operator, etc. Third, the
traditional chip defect identification method is difficult to achieve the multiple
requirements of intelligence, automation and high accuracy, low detection efficiency,
etc. [11-13].
Machine vision technology plays an important role in chip defect detection
technology. Features are mainly a combination of nondestructive testing, automation,
and intelligence. Not only is it safe and efficient, but also has high detection accuracy
[14]. Machine vision inspection technology is composed of three aspects: image
acquisition, software image processing, and image analysis [15-17]. The image
acquisition part is mainly to select a suitable light source, professional camera, and
lens to realize the picture acquisition of the sample. The principle is mainly to use the
image sensor to convert the light converged by the lens into an electrical signal, and
then into a digital signal, and pass it to the software processing part for analysis. The
software processing part mainly covers measures such as image denoising, image
enhancement, and edge detection [18-20]. The image analysis part includes three
parts: extraction of feature information, screening of effective features, and recognition
of images using classifiers. Mainly based on the extraction of effective feature
information from the pixels of the image, algorithms such as PCA are often used to
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
349
compress the image pixel data and reduce the high-dimensional image data to obtain
the features [21-22]. This makes it easier for the classifier to recognize feature
information during image recognition, thus improving the accuracy of classification
and more correct classification recognition [23-24].
Machine vision techniques are used in a large number of applications for defect
detection and target classification. In the literature [25], the location of the damage
appearance and identification of the damage type was roughly calculated using the
YOLOv3 detection network based on the image dataset of untouchable damage.
Using the designed level set algorithm, more accurate damage locations are obtained
in the image blocks. In the literature [26], a machine vision system based on a
recorder and image signal processing was proposed for automatic assessment. The
machine vision system consists of four modules, including a video acquisition module,
an image extraction module, an image processing module, and a trajectory state
evaluation module. Three classical edge detection methods were used and compared.
The literature [27] provides an overview of the application of machine vision models in
the field of fish classification and then discusses in detail the specific applications of
various classification methods. In addition, the challenges and future research
directions in the field of fish classification are discussed. The literature [28] reviews the
application of machine vision techniques for 3D dimensional and morphological
measurements of high-temperature metal components. In addition, two aspects are
described in detail, based on the principles and methods of measuring device
construction: laser scanning measurement and multi-view stereo vision techniques.
Through comparison and analysis, special attention is given to each method to
provide the necessary technical references for subsequent researchers. The literature
[29] presents a multi-defect stereo inspection system for magnetic rings based on
multi-camera vision technology to accomplish the automatic inspection of magnetic
rings. The system can simultaneously detect surface defects and measure ring height.
Two image processing algorithms are proposed, namely the image edge removal
algorithm (IERA) and the magnetic ring localization algorithm (MRLA). Based on these
two algorithms, a connected-domain filtering method for cracks, fibers, and large-area
defects is established to accomplish defect detection. The results show that the
system achieves a 99% recognition rate for defects such as cracks, adhesion, hanger
adhesion, and pitting. The literature [30] reviewed the principles, cameras, and
thermal data of infrared imaging-based machine vision and discussed the application
of deep learning in infrared imaging machine vision. Case studies of IR imaging-based
machine vision and deep learning on various platforms such as unmanned vehicles,
cell phones, and embedded systems are also reported. Machine vision techniques
have been rapidly developed in recent years for the detection of defects in high-
performance phase-noise measurement chips. By combing and analyzing chip defect
detection methods, the traditional machine vision problem of requiring different image
processing algorithms for classifying different tasks is solved, and the further
development of machine vision technology in the field of high-performance phase
noise measurement chip defect detection is promoted.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
350
compress the image pixel data and reduce the high-dimensional image data to obtain
the features [21-22]. This makes it easier for the classifier to recognize feature
information during image recognition, thus improving the accuracy of classification
and more correct classification recognition [23-24].
Machine vision techniques are used in a large number of applications for defect
detection and target classification. In the literature [25], the location of the damage
appearance and identification of the damage type was roughly calculated using the
YOLOv3 detection network based on the image dataset of untouchable damage.
Using the designed level set algorithm, more accurate damage locations are obtained
in the image blocks. In the literature [26], a machine vision system based on a
recorder and image signal processing was proposed for automatic assessment. The
machine vision system consists of four modules, including a video acquisition module,
an image extraction module, an image processing module, and a trajectory state
evaluation module. Three classical edge detection methods were used and compared.
The literature [27] provides an overview of the application of machine vision models in
the field of fish classification and then discusses in detail the specific applications of
various classification methods. In addition, the challenges and future research
directions in the field of fish classification are discussed. The literature [28] reviews the
application of machine vision techniques for 3D dimensional and morphological
measurements of high-temperature metal components. In addition, two aspects are
described in detail, based on the principles and methods of measuring device
construction: laser scanning measurement and multi-view stereo vision techniques.
Through comparison and analysis, special attention is given to each method to
provide the necessary technical references for subsequent researchers. The literature
[29] presents a multi-defect stereo inspection system for magnetic rings based on
multi-camera vision technology to accomplish the automatic inspection of magnetic
rings. The system can simultaneously detect surface defects and measure ring height.
Two image processing algorithms are proposed, namely the image edge removal
algorithm (IERA) and the magnetic ring localization algorithm (MRLA). Based on these
two algorithms, a connected-domain filtering method for cracks, fibers, and large-area
defects is established to accomplish defect detection. The results show that the
system achieves a 99% recognition rate for defects such as cracks, adhesion, hanger
adhesion, and pitting. The literature [30] reviewed the principles, cameras, and
thermal data of infrared imaging-based machine vision and discussed the application
of deep learning in infrared imaging machine vision. Case studies of IR imaging-based
machine vision and deep learning on various platforms such as unmanned vehicles,
cell phones, and embedded systems are also reported. Machine vision techniques
have been rapidly developed in recent years for the detection of defects in high-
performance phase-noise measurement chips. By combing and analyzing chip defect
detection methods, the traditional machine vision problem of requiring different image
processing algorithms for classifying different tasks is solved, and the further
development of machine vision technology in the field of high-performance phase
noise measurement chip defect detection is promoted.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
The environmental problems caused by chip defects have been a hot issue of great
concern. To improve the correct rate and detection rate of high-performance phase
noise measurement chip defect detection, and improve the ecological environment
pollution caused by the chip production process. In this paper, based on the ecological
environments problems such as waste gas, wastewater, and solid waste caused by
chip defects in the chip production process, we build a machine vision inspection
technology for detecting defects contained in high-performance phase noise
measurement chips using three aspects: image acquisition, software image
processing, and image analysis.
2. ENVIRONMENTAL POLLUTION CAUSED BY CHIP
DEFECTS
The environmental pollution caused by chip defects mainly comes from the three
wastes generated in the manufacturing process, namely wastewater, exhaust gas,
and solid waste. The exhaust gas, wastewater, and solid waste generated during the
chip manufacturing process contain high concentrations of organic pollutants as well
as fluorinated pollutants and other pollution factors that are seriously harmful to the
environment. A large number of gases are produced during the manufacturing process
of chips due to defects. For example, PH3, BF3, Cl2, SF6, CF4, C4F8, BCl3 and other
organic substances, ammonia nitrogen, fluorine-containing pollutants, and other
factors. If these gases are directly discharged without treatment, they will cause great
pollution to the environment and direct harm to human health.
With the continuous improvement of air pollution control, the concentration of flue
gas emissions is becoming more and more strict, and as far as the enterprises
themselves are concerned, the urgency of waste gas treatment is much greater than
that of wastewater treatment. The more harmful components of the exhaust gas
produced in the manufacturing process are hydrogen sulfide, fluorinated dust, and
sulfur dioxide. These harmful substances can cause acute and chronic diseases of
the human respiratory system. The wastewater produced in the chip manufacturing
process contains a large amount of organic pollutants. Once these pollutants enter the
surrounding water bodies, they will cause the microorganisms in the water bodies to
multiply rapidly. This causes a dramatic decrease in dissolved oxygen in the water,
which leads to the death of aquatic organisms in the water body due to lack of
oxygen. Ammonia nitrogen is an important nutrient in the environment of water bodies,
and its random discharge will lead to eutrophication of water bodies. At the same time,
ammonia nitrogen is also a major oxygen-consuming pollutant, and when dissolved
oxygen reacts with ammonia nitrogen in the water body, it will bring great toxic effects
to other kinds of aquatic organisms in the water body. In addition, if ammonia nitrogen
is ingested by the human body for a long time through the food chain, it will be
transformed into ammonium nitrite in the human body under specific conditions, and
the long-term accumulation of this substance in the human body will make the risk of
cancer rise sharply. Fluorine is an essential element for the human body and is one of
the main components of human bones and teeth. It also plays an important role in the
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
351
formation of bone tissue and tooth enamel and participates in metabolic processes by
activating or inhibiting the activity of various enzymes. Lack of fluoride in the body or
excessive fluoride inhalation can have serious health consequences. The cavities we
usually see formed by erosion of tooth enamel, muscle atrophy, thickening of joints,
and other large bony conditions are all unfavorable lesions due to the lack of fluoride
in the body. Excess fluoride, on the other hand, can likewise bring about unfavorable
some lesions, including bone deformation, back and leg pain, loss of labor force, and
even worse, death. Fluorine is not only harmful to humans but also to the natural
environment we live in. Some plants absorb some fluorine dissolved in water through
the soil, and the excess fluorine can have a serious toxic effect on plants, which are
eventually eaten by animals and humans.
2.1. EXHAUST GAS AND SOLID WASTE GENERATION DUE
TO CHIP DEFECTS
Acid exhaust gas mainly comes from the exhaust gas generated by the
volatilization of acidic raw materials from the wet etching of the array production
process and the medium gas periodically emitted from the excimer laser annealing
device. The main pollutants are acid mist such as HCL and fluoride. The alkaline
exhaust gas mainly comes from the alkaline volatile substances generated from the
development process of the array and color film production process, and the main
pollutants are NH3. dry etching and chemical vapor deposition exhaust gas mainly
come from some gas raw materials in the dry etching process to produce a certain
amount of reaction exhaust gas and unreacted raw material gas. The main pollutants
are N2O, SiH4, NH3, NF3, PH3, BF3, Cl2, SF6, CF4, C4F8, BCl3 and reaction waste
gas fluoride and chloride, etc. Organic waste gas mainly comes from the volatilization
of organic gases from organic raw materials in the production process, and the main
pollutants are non-methane total hydrocarbons. Solid waste mainly comes from the
waste of various raw materials used in the production process of array, color film,
organic vapor deposition, and box formation. This includes their packaging containers,
organic wiping materials, residual liquids, and expired unusable materials. Chip
manufacturing waste gas as well as solid waste generation is shown in Table 1.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
352
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
Table 1. Generation of chip defect waste gas and solid waste
2.2. CHIP DEFECT WASTEWATER GENERATION
The main pollutants of acid-base wastewater are pH, COD, BOD5, SS, F-, NH3-N.
The fluorinated wastewater mainly comes from the PECVD deposition and dry etching
process waste gas POU purification system discharge, and the array process
fluorinated cleaning wastewater. purification device for treatment. The main pollutants
in the discharged wastewater are F-, pH, COD, BOD5, SS, NH3-N. The phosphorus-
containing wastewater mainly comes from wet etching using phosphoric acid, sulfuric
acid, nitric acid, and other raw materials. Organic wastewater mainly comes from the
production process of film, organic vapor deposition, box formation, etc., glue
Pollution
category
Pollution
source
Major
pollutants
Emission
method
Production
(t/a)
Yield
concentration
(mg/m3)
Acid
waste
gas
Acidic volatile
gases from array
wet etching
Fluoride
Continuous
28.26 40
NOX 83.79 120
Alkaline
waste
gas
Volatile alkaline
gases emitted
from developing
production
processes such
as arrays and
color filters
NH3
Continuous
20.56 65
HCL 118.02 350
Fluoride 201.7 520
Dry
etching
and
chemical
vapor
depositio
n exhaust
Unreacted gas
and reaction
waste gas
discharged from
vapor
deposition, dry
etching, doping,
etc.
NOX
Continuous
48.01 128
SO2 37.67 1500
Cl2 198.70 600
Organic
waste
gas
Organic waste
gas generated
by coating,
peeling,
evaporation, etc.
from arrays,
color filters,
organic
evaporation, and
box formation
Total non-
methane
hydrocarbon
s
Continuous 5968.0 6000
Solid
waste
Waste reagent
containers,
organic wiping
materials,
expired raw
materials,
residual liquids
Chemical
reagents
containing
acids, bases,
alcohol
esters,
ethers, etc.
Regular 8000
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
353
application, glue stripping, detergent cleaning, etc. The production processes all use
organic materials as raw materials and ultra-pure water is used for cleaning after the
processes. Therefore, organic wastewater mainly comes from the cleaning
wastewater of production processes such as array, color film, organic vapor
deposition, and box formation, and organic wastewater is discharged from alkaline
and etching exhaust gas washing and purification system. The main pollutants are
COD, BOD5, SS, NH3-N, pH, etc. The wastewater generation situation of the chip
manufacturing industry is shown in Table 2.
Table 2. Wastewater generation of chip defect
3. APPLICATION OF MACHINE VISION-BASED
TECHNOLOGY FOR HIGH-PERFORMANCE PHASE
NOISE MEASUREMENT CHIP DEFECT DETECTION
Machine vision technology plays an important role in chip defect detection.
Features are mainly non-destructive testing, automation, and intelligence combined,
not only good safety, and high efficiency, but also high detection accuracy.
Machine vision inspection technology is composed of three aspects: image
acquisition, software image processing, and image analysis. The image acquisition
part is mainly to choose the appropriate light source, professional cameras, and
lenses, to achieve the picture acquisition of chip samples. The principle is mainly to
use the image sensor to convert the light converged by the lens into electrical signals,
and then into digital signals, and pass to the software processing part for analysis.
The software processing part mainly covers measures such as image denoising,
Pollution
category Pollution source Major pollutants Emission
method
Wastew
ater
volume(
m
3
/d)
Acidic
wastewater
Acid-base waste gas
treatment system
discharge waste water
pure water preparation
of waste water
PH, COD, BOD5,
SS、F－, NH3-N Continuous 4202
Fluorinated
wastewater
Wastewater from
exhaust gas
purification systems for
ECVD deposition and
dry etching processes
PH, COD, BOD5,
SS、F－, NH3-N Continuous 2518
Phosphorus-
containing
wastewater
Array wet etching
discharge cleaning
wastewater
PH, COD, BOD5,
SS, Phosphate Continuous 683
Organic
wastewater
Equipment circulating
cooling water system
sewage, boiler
sewage, etc.
Continuous 5496
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
354
application, glue stripping, detergent cleaning, etc. The production processes all use
organic materials as raw materials and ultra-pure water is used for cleaning after the
processes. Therefore, organic wastewater mainly comes from the cleaning
wastewater of production processes such as array, color film, organic vapor
deposition, and box formation, and organic wastewater is discharged from alkaline
and etching exhaust gas washing and purification system. The main pollutants are
COD, BOD5, SS, NH3-N, pH, etc. The wastewater generation situation of the chip
manufacturing industry is shown in Table 2.
Table 2. Wastewater generation of chip defect
3. APPLICATION OF MACHINE VISION-BASED
TECHNOLOGY FOR HIGH-PERFORMANCE PHASE
NOISE MEASUREMENT CHIP DEFECT DETECTION
Machine vision technology plays an important role in chip defect detection.
Features are mainly non-destructive testing, automation, and intelligence combined,
not only good safety, and high efficiency, but also high detection accuracy.
Machine vision inspection technology is composed of three aspects: image
acquisition, software image processing, and image analysis. The image acquisition
part is mainly to choose the appropriate light source, professional cameras, and
lenses, to achieve the picture acquisition of chip samples. The principle is mainly to
use the image sensor to convert the light converged by the lens into electrical signals,
and then into digital signals, and pass to the software processing part for analysis.
The software processing part mainly covers measures such as image denoising,
Pollution
category
Pollution source
Major pollutants
Emission
method
Wastew
ater
volume(
m3/d)
Acidic
wastewater
Acid-base waste gas
treatment system
discharge waste water
pure water preparation
of waste water
PH, COD, BOD5,
SS、F－, NH3-N
Continuous
4202
Fluorinated
wastewater
Wastewater from
exhaust gas
purification systems for
ECVD deposition and
dry etching processes
PH, COD, BOD5,
SS、F－, NH3-N
Continuous
2518
Phosphorus-
containing
wastewater
Array wet etching
discharge cleaning
wastewater
PH, COD, BOD5,
SS, Phosphate
Continuous
683
Organic
wastewater
Equipment circulating
cooling water system
sewage, boiler
sewage, etc.
Continuous
5496
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
image enhancement, and edge detection. The image analysis part includes three
parts: extraction of feature information, screening of effective features, and recognition
of images using classifiers. It is mainly based on the extraction of effective feature
information from the pixels of the image. Algorithms such as PCA are often used to
compress the image pixel data and reduce the high-dimensional image data to obtain
features. This makes it easier for the classifier to recognize feature information during
image recognition, thus improving the accuracy of classification and more correct
classification recognition. The detection workflow diagram is shown in Figure 1.
Figure 1. Detection workflow diagram
4. RESULTS AND ANALYSIS
Machine vision technology for chip defect identification detection has the
advantages of automation, high detection efficiency, and high accuracy of defect
identification. With the rapid development of machine vision technology in recent
years, machine vision inspection technology has been used in a large number of
applications in chip defect detection. Selected 100 high-performance phase noise
measurement chips containing different kinds of defects, through the application of
machine vision technology detection, two cases of false detection occurred. Once a
foreign object on the chip surface was not detected and once a metallic contaminant
present on the chip was not detected. This resulted in a 98% correct rate of chip
defect detection. By fixing the chip defect problem, the environmental problems
caused by chip defects were improved in three ways.
1. Acidic emissions from the manufacturing process due to chip defects mainly
come from the volatilization of acidic raw materials from the wet etching
process of the array manufacturing process, and the periodic emission of
dielectric gases from the excimer laser annealing unit. The main pollutant is
"fluoride". The alkaline exhaust gas mainly comes from the alkaline volatile
substances generated from the development process of the array and color
film production process. The main pollutant is "NH3". Dry etching and chemical
vapor deposition waste gas mainly come from some gas raw materials in the
dry etching process to produce a certain amount of reaction waste gas and
unreacted raw material gas. The main pollutant is "NOX". Organic waste gas
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
355
mainly comes from the volatilization of organic gases in the organic raw
materials during the production process, and the main pollutant is non-total
methane hydrocarbon. The application of machine vision technology in the
detection of defects of chips with high-performance phase noise measurement
has improved the defects of chips and reduced the amount of acidic exhaust
gas "fluoride" from 28.26t/a to 20.53t/a. The amount of "NH3", the main
pollutant of dry etching and chemical vapor deposition waste gas, is reduced
from 48.01t/a to 35.23t/a. The amount of "NOX", the main pollutant of organic
waste gas, is reduced from 20.56t/a to 15.2t/a. The amount of "non-methane
total hydrocarbons", the main pollutant of organic waste gas, was reduced from
5968.0t/a to 4000t/a.
2.
The wastewater generated in the manufacturing process due to chip defects
mainly includes acid and alkali wastewater, phosphorus-containing wastewater,
fluorine-containing wastewater, and organic wastewater. Acid-base wastewater
mainly comes from acid-base cleaning wastewater discharged from the
production system, acid exhaust gas scrubbing and purification system
discharge wastewater, and regeneration backwash wastewater discharged
from the pure water preparation system. Fluorine-containing wastewater mainly
comes from wastewater discharged from PECVD deposition and dry etching
process exhaust gas POU purification system, and fluorine-containing cleaning
wastewater from array process. Phosphorus-containing wastewater mainly
comes from wet etching using phosphoric acid, sulfuric acid, nitric acid, and
other raw materials, and the cleaning wastewater after etching mainly contains
pH, COD, BOD5, SS, phosphate, and other pollutants. Organic wastewater
mainly comes from the array, color film, organic vapor deposition, box
formation, and other production process cleaning wastewater, and organic
wastewater discharged from alkaline and etching exhaust gas scrubbing and
purification systems. Through the application of machine vision technology in
the detection of defects of chips with high-performance phase noise
measurement, the defects of chips are improved, so that the wastewater
volume of acidic wastewater from the chip manufacturing process is reduced
from 4202m3/d to 3500m3
/d. The wastewater volume of phosphorus-containing
wastewater is reduced from 683m3/d to 552m3
/d. The wastewater volume of
fluorine-containing wastewater is reduced from 2518m3/d to 2208m3
/d. The
wastewater volume of organic wastewater is reduced from 5496m3
/d to
4600m3/d.
3.
The solid waste generated in the manufacturing process due to chip defects
mainly comes from the waste of various raw materials used in the production
processes of the array, color film, organic vapor deposition, and box formation.
This includes their packaging containers, organic wiping materials, residual
liquids, expired unusable materials, etc. Through the application of machine
vision technology in the detection of defects in high-performance phase noise
measurement chips, the defects of chips have been improved and the amount
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
356
mainly comes from the volatilization of organic gases in the organic raw
materials during the production process, and the main pollutant is non-total
methane hydrocarbon. The application of machine vision technology in the
detection of defects of chips with high-performance phase noise measurement
has improved the defects of chips and reduced the amount of acidic exhaust
gas "fluoride" from 28.26t/a to 20.53t/a. The amount of "NH3", the main
pollutant of dry etching and chemical vapor deposition waste gas, is reduced
from 48.01t/a to 35.23t/a. The amount of "NOX", the main pollutant of organic
waste gas, is reduced from 20.56t/a to 15.2t/a. The amount of "non-methane
total hydrocarbons", the main pollutant of organic waste gas, was reduced from
5968.0t/a to 4000t/a.
2. The wastewater generated in the manufacturing process due to chip defects
mainly includes acid and alkali wastewater, phosphorus-containing wastewater,
fluorine-containing wastewater, and organic wastewater. Acid-base wastewater
mainly comes from acid-base cleaning wastewater discharged from the
production system, acid exhaust gas scrubbing and purification system
discharge wastewater, and regeneration backwash wastewater discharged
from the pure water preparation system. Fluorine-containing wastewater mainly
comes from wastewater discharged from PECVD deposition and dry etching
process exhaust gas POU purification system, and fluorine-containing cleaning
wastewater from array process. Phosphorus-containing wastewater mainly
comes from wet etching using phosphoric acid, sulfuric acid, nitric acid, and
other raw materials, and the cleaning wastewater after etching mainly contains
pH, COD, BOD5, SS, phosphate, and other pollutants. Organic wastewater
mainly comes from the array, color film, organic vapor deposition, box
formation, and other production process cleaning wastewater, and organic
wastewater discharged from alkaline and etching exhaust gas scrubbing and
purification systems. Through the application of machine vision technology in
the detection of defects of chips with high-performance phase noise
measurement, the defects of chips are improved, so that the wastewater
volume of acidic wastewater from the chip manufacturing process is reduced
from 4202m3/d to 3500m3/d. The wastewater volume of phosphorus-containing
wastewater is reduced from 683m3/d to 552m3/d. The wastewater volume of
fluorine-containing wastewater is reduced from 2518m3/d to 2208m3/d. The
wastewater volume of organic wastewater is reduced from 5496m3/d to
4600m3/d.
3. The solid waste generated in the manufacturing process due to chip defects
mainly comes from the waste of various raw materials used in the production
processes of the array, color film, organic vapor deposition, and box formation.
This includes their packaging containers, organic wiping materials, residual
liquids, expired unusable materials, etc. Through the application of machine
vision technology in the detection of defects in high-performance phase noise
measurement chips, the defects of chips have been improved and the amount
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
of solid waste generated from the chip manufacturing process has been
reduced from 8000t/a to 6500t/a. See Figure 2 for details.
(a) Organic waste gas, wastewater, and solid waste
(b) Acid and alkaline waste gas, dry etching, and chemical vapor deposition waste gas
Figure 2. Quantification of waste generation
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
357
Inspection using machine vision technology is an important step in the chip
production process. Geometric measurement is the key technology for automatic
inspection in traditional automated inspection. Although this technology can achieve
automatic inspection, in terms of detection accuracy, speed is relatively poor, and has
gradually failed to meet the automation efficiency to further enhance the requirements.
The application of machine vision technology in automatic inspection can use CT,
laser scanning, and other technologies to synchronize automatic inspection, which
can not only effectively improve the speed of automatic inspection, but also further
improve the accuracy of automatic inspection. Secondly, the biggest advantage of
machine vision technology in automatic inspection is that automatic inspection
technology can realize the detection of chip appearance. The traditional geometric
inspection method cannot realize the processing of chip appearance information and
can only deal with geometric dimensions. Machine vision technology, on the other
hand, can effectively realize the processing of the chip surface, thus further enhancing
the accuracy of the detection. Based on the generation and emission of major
pollutants before and after chip defect detection, it can be seen that the use of
machine vision technology to detect chip defects not only helps to repair chip defects
promptly but also greatly reduces the amount of waste gas, wastewater and solid
pollutants generated by chip defects, alleviating the degree of harm caused to the
surrounding environment. In summary, machine vision-based chip defect detection
technology is of great importance to improve the ecological level.
5. DISCUSSION
Machine vision technology is not only the core of artificial intelligence technology,
but also has a very important significance for the current social production efficiency
improvement. In the future machine vision technology is bound to achieve further
development. Extraction of chip defect information characteristics, not only can
achieve high precision detection of high-performance noise measurement chip defects
but also can directly achieve end-to-end chip defect detection, reducing the
complexity of engineering. Also, the scope of application in social production will be
further promoted. Most of the current methods for chip defect detection are based on
two-dimensional images. Such methods can only obtain limited flat feature
information, and cannot obtain the spatial feature information of material defects.
Therefore, how to capture and use three-dimensional defect information more
accurately to detect defects is also a direction worthy of future research.
6. CONCLUSION
This paper is based on the ecological environmental problems caused by chip
defects in the chip production process. Choose the appropriate light source,
professional camera, and lens to achieve the picture acquisition of chip samples.
Through image denoising, image enhancement, edge detection, and other measures,
image processing is completed. After three parts: extraction of feature information,
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
358
Inspection using machine vision technology is an important step in the chip
production process. Geometric measurement is the key technology for automatic
inspection in traditional automated inspection. Although this technology can achieve
automatic inspection, in terms of detection accuracy, speed is relatively poor, and has
gradually failed to meet the automation efficiency to further enhance the requirements.
The application of machine vision technology in automatic inspection can use CT,
laser scanning, and other technologies to synchronize automatic inspection, which
can not only effectively improve the speed of automatic inspection, but also further
improve the accuracy of automatic inspection. Secondly, the biggest advantage of
machine vision technology in automatic inspection is that automatic inspection
technology can realize the detection of chip appearance. The traditional geometric
inspection method cannot realize the processing of chip appearance information and
can only deal with geometric dimensions. Machine vision technology, on the other
hand, can effectively realize the processing of the chip surface, thus further enhancing
the accuracy of the detection. Based on the generation and emission of major
pollutants before and after chip defect detection, it can be seen that the use of
machine vision technology to detect chip defects not only helps to repair chip defects
promptly but also greatly reduces the amount of waste gas, wastewater and solid
pollutants generated by chip defects, alleviating the degree of harm caused to the
surrounding environment. In summary, machine vision-based chip defect detection
technology is of great importance to improve the ecological level.
5. DISCUSSION
Machine vision technology is not only the core of artificial intelligence technology,
but also has a very important significance for the current social production efficiency
improvement. In the future machine vision technology is bound to achieve further
development. Extraction of chip defect information characteristics, not only can
achieve high precision detection of high-performance noise measurement chip defects
but also can directly achieve end-to-end chip defect detection, reducing the
complexity of engineering. Also, the scope of application in social production will be
further promoted. Most of the current methods for chip defect detection are based on
two-dimensional images. Such methods can only obtain limited flat feature
information, and cannot obtain the spatial feature information of material defects.
Therefore, how to capture and use three-dimensional defect information more
accurately to detect defects is also a direction worthy of future research.
6. CONCLUSION
This paper is based on the ecological environmental problems caused by chip
defects in the chip production process. Choose the appropriate light source,
professional camera, and lens to achieve the picture acquisition of chip samples.
Through image denoising, image enhancement, edge detection, and other measures,
image processing is completed. After three parts: extraction of feature information,
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
screening of effective features, and recognition of images using classifiers, high-
performance phase noise measurement chip defects are detected. The conclusions of
the obtained study are as follows.
1.
Through the comparison of traditional chip defect detection technology and
machine vision technology, it is concluded that machine vision technology for
chip defect identification and detection has the advantages of automation, high
detection efficiency, and high accuracy of defect identification. The biggest
advantage of machine vision technology in automated inspection is that the
automatic detection technology can achieve the detection of chip appearance.
Traditional geometric inspection methods can not achieve the processing of
chip appearance information, only geometric size, while machine vision
technology can effectively achieve the processing of the chip surface, thereby
further enhancing the accuracy of detection.
2.
Chip defect identification detection by machine vision technology has the
advantages of automation, high detection efficiency, and high accuracy of
defect identification. Machine vision inspection technology has been applied in
a large number of chip defect detection, and the correct rate of chip defect
detection is as high as 98% through the application of machine vision
technology detection.
3. Analysis of the type of environmental pollution caused by chip defects, and the
use of machine vision technology for chip defect detection. By dealing with chip
defects promptly, the amount of waste gas, wastewater, and solid waste
generated is reduced. The amount of organic waste gas generated was
reduced from 5968.0t/a to 4000t/a, the amount of organic wastewater
generated was reduced from 5496m3/d to 4600m3
/d, and the amount of solid
waste generated was reduced from 8000t/a to 6500t/a. The ecological
environment around the enterprise was improved.
REFERENCES
(1)
Wu, J. (2021). Chip Appearance Defect Recognition Based on Convolutional
Neural Network. Sensors, 21.
(2)
Larouci, B., Ayad, A. N. E. I., Alharbi, H., et al. (2022). Investigation on New
Metaheuristic Algorithms for Solving Dynamic Combined Economic
Environmental Dispatch Problems. Sustainability, 14.
(3) Li, C., Chen, H., Li, X., et al. (2020). A review for cervical histopathology image
analysis using machine vision approaches. Artificial Intelligence Review, 53(1).
(4)
Zhang, X., Gong, W., Xu, X. (2020). Magnetic Ring Multi-Defect Stereo
Detection System Based on Multi-Camera Vision Technology. Sensors, 20(2),
392.
(5)
Choi, S. M., Jeong, J. C., Kim, J., et al. (2020). A novel three-dimensional
magnetic particle imaging system based on the frequency mixing for the point-of-
care diagnostics. Scientific Reports, 10(1).
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
359
(6)
Wang, D., Sun, B. C., Wang, J. X., et al. (2020). Can Masks Be Reused After
Hot Water Decontamination During the COVID-19 Pandemic? Engineering.
(7)
Huang, Y. (2021). Flaw Detection in Highly Scattering Materials Using a Simple
Ultrasonic Sensor Employing Adaptive Template Matching. Sensors, 22.
(8)
Liu, E., Liu, Y., Chen, Y., et al. (2022). Measurement method of bolt hole
assembly stress based on the combination of ultrasonic longitudinal and
transverse waves. Applied Acoustics, 189, 108603.
(9)
Anderson, D. L., Cunningham, W. C. (2020). Nondestructive Determination of
Lead, Cadmium, Tin, Antimony, and Barium in Ceramic Glazes by Radioisotope
X-Ray Fluorescence Spectrometry. Journal of AOAC International, 5.
(10)
Li, D., Wang, Q., Li, X., et al. (2022). Recent advances of machine vision
technology in fish classification. ICES Journal of Marine Science, 2.
(11)
Tian, Q., Conceptualization, Methodology, Data Curation, Software, Formal
analysis, Writing - original draft, Writing - review & editing, Analysis C. F.,
Conceptualization methodology Q. F., et al. (2021). Improved cortical surface
reconstruction using sub-millimeter resolution MPRAGE by image denoising.
NeuroImage.
(12) Yang, W., Wang, S., Fang, Y., et al. (2021). Band Representation-Based Semi-
Supervised Low-Light Image Enhancement: Bridging the Gap Between Signal
Fidelity and Perceptual Quality. IEEE Transactions on Image Processing.
(13)
Lu, S., Liu, Z., Chen, Y., et al. (2021). A novel subpixel edge detection method of
pantograph slide in complicated surroundings. IEEE Transactions on Industrial
Electronics, PP(99), 1-1.
(14)
Tran, N., Wang, C. C. (2021). Enhancement of the accuracy of ultrasonic
flowmeters by applying the PCA algorithm in predicting flow patterns.
Measurement Science and Technology, 32(8), 085901 (18pp).
(15)
Jones, I., Zhu, M., Zhang, J., et al. (2021). The application of spent tyre activated
carbons as low-cost environmental pollution adsorbents: A technical review.
Journal of Cleaner Production, 312, 127566.
(16)
Xie, C., Liu, J., Liang, J., et al. (2021). Optimizing environmental pollution
controls in response to textile dyeing sludge, incineration temperature, CaO
conditioner, and ash minerals. Science of The Total Environment, 785, 147219.
(17)
Ciobanu, C., Tudor, P., Istrate, I. A., et al. (2022). Assessment of Environmental
Pollution in Cement Plant Areas in Romania by Co-Processing Waste in Clinker
Kilns. Energies, 15.
(18)
Liu, D., Song, C., Fang, C., et al. (2021). A recommended nitrogen application
strategy for high crop yield and low environmental pollution at a basin scale.
Science of The Total Environment, 792(26), 148464.
(19)
Zandalinas, S. I., Fritschi, F. B., Mittler, R. (2021). Global Warming, Climate
Change, and Environmental Pollution: Recipe for a Multifactorial Stress
Combination Disaster. Trends in Plant Science.
(20)
Torres, F. G., De-La-Torre, G. E. (2021). Environmental pollution with antifouling
paint particles: Distribution, ecotoxicology, and sustainable alternatives. Marine
Pollution Bulletin, 169, 112529.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
360
(6) Wang, D., Sun, B. C., Wang, J. X., et al. (2020). Can Masks Be Reused After
Hot Water Decontamination During the COVID-19 Pandemic? Engineering.
(7) Huang, Y. (2021). Flaw Detection in Highly Scattering Materials Using a Simple
Ultrasonic Sensor Employing Adaptive Template Matching. Sensors, 22.
(8) Liu, E., Liu, Y., Chen, Y., et al. (2022). Measurement method of bolt hole
assembly stress based on the combination of ultrasonic longitudinal and
transverse waves. Applied Acoustics, 189, 108603.
(9) Anderson, D. L., Cunningham, W. C. (2020). Nondestructive Determination of
Lead, Cadmium, Tin, Antimony, and Barium in Ceramic Glazes by Radioisotope
X-Ray Fluorescence Spectrometry. Journal of AOAC International, 5.
(10) Li, D., Wang, Q., Li, X., et al. (2022). Recent advances of machine vision
technology in fish classification. ICES Journal of Marine Science, 2.
(11) Tian, Q., Conceptualization, Methodology, Data Curation, Software, Formal
analysis, Writing - original draft, Writing - review & editing, Analysis C. F.,
Conceptualization methodology Q. F., et al. (2021). Improved cortical surface
reconstruction using sub-millimeter resolution MPRAGE by image denoising.
NeuroImage.
(12) Yang, W., Wang, S., Fang, Y., et al. (2021). Band Representation-Based Semi-
Supervised Low-Light Image Enhancement: Bridging the Gap Between Signal
Fidelity and Perceptual Quality. IEEE Transactions on Image Processing.
(13) Lu, S., Liu, Z., Chen, Y., et al. (2021). A novel subpixel edge detection method of
pantograph slide in complicated surroundings. IEEE Transactions on Industrial
Electronics, PP(99), 1-1.
(14) Tran, N., Wang, C. C. (2021). Enhancement of the accuracy of ultrasonic
flowmeters by applying the PCA algorithm in predicting flow patterns.
Measurement Science and Technology, 32(8), 085901 (18pp).
(15) Jones, I., Zhu, M., Zhang, J., et al. (2021). The application of spent tyre activated
carbons as low-cost environmental pollution adsorbents: A technical review.
Journal of Cleaner Production, 312, 127566.
(16) Xie, C., Liu, J., Liang, J., et al. (2021). Optimizing environmental pollution
controls in response to textile dyeing sludge, incineration temperature, CaO
conditioner, and ash minerals. Science of The Total Environment, 785, 147219.
(17) Ciobanu, C., Tudor, P., Istrate, I. A., et al. (2022). Assessment of Environmental
Pollution in Cement Plant Areas in Romania by Co-Processing Waste in Clinker
Kilns. Energies, 15.
(18) Liu, D., Song, C., Fang, C., et al. (2021). A recommended nitrogen application
strategy for high crop yield and low environmental pollution at a basin scale.
Science of The Total Environment, 792(26), 148464.
(19) Zandalinas, S. I., Fritschi, F. B., Mittler, R. (2021). Global Warming, Climate
Change, and Environmental Pollution: Recipe for a Multifactorial Stress
Combination Disaster. Trends in Plant Science.
(20) Torres, F. G., De-La-Torre, G. E. (2021). Environmental pollution with antifouling
paint particles: Distribution, ecotoxicology, and sustainable alternatives. Marine
Pollution Bulletin, 169, 112529.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
(21)
Liu, Z., Wang, Z., Chen, H., et al. (2021). Hydrochar and pyrochar for sorption of
pollutants in wastewater and exhaust gas: A critical review. Environmental
Pollution, 268(Pt B), 115910.
(22) Pavlenko, A. (2021). Capture of Pollutants from Exhaust Gases by Low-
Temperature Heating Surfaces. Energies, 15.
(23)
Chao, S., Guo, L., Sun, S. (2021). Zombie problem: Normal firms' wastewater
pollution. Journal of Cleaner Production, 330(4), 129893.
(24)
Cheng, D., Ngo, H. H., Guo, W., et al. (2020). A critical review on antibiotics and
hormones in swine wastewater: Water pollution problems and control
approaches. Journal of Hazardous Materials, 387, 121682-.
(25)
Tao, J., Chen, Q., Xu, J., et al. (2022). Utilization of Both Machine Vision and
Robotics Technologies in Assisting Quality Inspection and Testing. Mathematical
Problems in Engineering, 2022.
(26)
Wang, S. M., Liao, C. L., Ni, Y. Q. (2020). A Machine Vision System based on
Driving Recorder for Automatic Inspection of Rail Curvature. IEEE Sensors
Journal, PP(99), 1-1.
(27)
Li, D., Wang, Q., Li, X., et al. (2022). Recent advances of machine vision
technology in fish classification. ICES Journal of Marine Science, 2022(2), 2.
(28)
Li, Feng & Abo Keir, Mohammed Yousuf. (2021). Mathematical model of back
propagation for stock price forecasting. Applied Mathematics and Nonlinear
Sciences. doi:10.2478/AMNS.2021.1.00089.
(29)
Zhang, X., Gong, W., Xu, X. (2020). Magnetic Ring Multi-Defect Stereo
Detection System Based on Multi-Camera Vision Technology. Sensors, 20(2),
392.
(30)
He, Y., Deng, B., Wang, H., et al. (2021). Infrared machine vision and infrared
thermography with deep learning: A review. Infrared Physics & Technology,
116(2), 103754.
(31)
Shang, Y., You, B., Shang, L. (2016). China’s environmental strategy towards
reducing deep groundwater exploitation. Environmental Earth Sciences, 75,
1439.
(32)
Zhang, J., Shang, Y., Cui, M., Luo, Q., Zhang, R. (2022). Successful and
sustainable governance of the lower Yellow River, China: A floodplain utilization
approach for balancing ecological conservation and development. Environment,
Development and Sustainability, 24, 3014-3038.
(33)
Lu, S. B., Shang, Y. Z.*, Pei, L., Li, W., Wu, X. H. (2017). The effects of rural
domestic sewage reclaimed water drip irrigation on characteristics of
rhizosphere soil. Applied Ecology and Environmental Research, 15(4),
1145-1155.
(34)
Wang, J., Liu, J., Shang, Y., Jiang, D., Xiao, W. (2015). China’s campaign to
create artificial water surfaces in drought-affected regions must consider
prevention measures for ecological problems. Environmental Earth Sciences,
74(6), 5457-5462.
(35)
Lu, S., Shang, Y., Zhang, H. (2020). Evaluation on Early Drought Warning
System in the Jinghui Channel Irrigation Area. International Journal of
Environmental Research and Public Health, 17(1), 374.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
361
(36)
Panagopoulos, A., Haralambous, K. J. (2020). Environmental impacts of
desalination and brine treatment - Challenges and mitigation measures. Marine
Pollution Bulletin, 161(Pt B), 111773.
(37)
Wang, T., Wang, P., Cai, S., et al. (2020). A Unified Trustworthy Environment
Establishment Based on Edge Computing in Industrial IoT. IEEE Transactions on
Industrial Informatics, 16(9), 6083-6091.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.347-362
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
362