Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976169
Bo Cao, L. Yao
In this paper, based on a grain processing device, a bilinear stochastic distribution system (SDS) is established based on its input and output data. The problem of fault diagnosis (FD) and for the bilinear stochastic distribution system when the actuator fault is studied. A new unknown input observer (UIO) is designed to diagnose the fault. A simulation example is given to verify the proposed algorithm.
{"title":"Fault diagnosis for bilinear stochastic distribution systems with actuator fault","authors":"Bo Cao, L. Yao","doi":"10.1109/INDIN51773.2022.9976169","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976169","url":null,"abstract":"In this paper, based on a grain processing device, a bilinear stochastic distribution system (SDS) is established based on its input and output data. The problem of fault diagnosis (FD) and for the bilinear stochastic distribution system when the actuator fault is studied. A new unknown input observer (UIO) is designed to diagnose the fault. A simulation example is given to verify the proposed algorithm.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128163434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976110
Wenrui Zhao, Chengyi Pu
The relationship between risk and asset returns is an important basis for investment decision.The mystery of "idiosyncratic volatility" shows that this relationship is still unclear.How to measure the correlation of risk and returns accurately has always been a popular investment spot. Traditional research of tail risk from one-dimensional and multi-dimensional perspective is relatively rich, using conditional heteroscedasticity model or extreme value theory to measure the basic risk indicators such as Value at Risk(VaR) and Expected Shortfall(ES),or estimating the common tail risk factor based on cross-sectional data of stocks.Existing research does not consider the same direction changes between asset and market returns,which is more pronounced during market crashes.In this paper, we examine the impact of mixed tail risk on the expected stock returns from multi-dimensional perspective based on coupla method.We find that: (1) The coefficient of lower tail dependence(LTD) can capture market crashes,we can use LTD as an warning indicator for market crashes. Stocks traded on Shenzhen Main Board with strong LTD have higher future returns than that with weaker LTD, but this conclusion does not apply to the stocks traded on Small and Mid Enterprise board(SME board) and Growth Enterprise market(GEM). (2) In the period of financial crisis, the positive impact of stock mixed tail risk on stock expected return will be significantly enhanced.High circulation market capitalization and high turnover rate can reduce this impact. (3) Non-tradable Share Reform increases the liquidity of stocks,reducing the risk premium of mixed tail risk.
{"title":"Study on the Relationship between Mixed Tail Risk and Expected Stock Returns","authors":"Wenrui Zhao, Chengyi Pu","doi":"10.1109/INDIN51773.2022.9976110","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976110","url":null,"abstract":"The relationship between risk and asset returns is an important basis for investment decision.The mystery of \"idiosyncratic volatility\" shows that this relationship is still unclear.How to measure the correlation of risk and returns accurately has always been a popular investment spot. Traditional research of tail risk from one-dimensional and multi-dimensional perspective is relatively rich, using conditional heteroscedasticity model or extreme value theory to measure the basic risk indicators such as Value at Risk(VaR) and Expected Shortfall(ES),or estimating the common tail risk factor based on cross-sectional data of stocks.Existing research does not consider the same direction changes between asset and market returns,which is more pronounced during market crashes.In this paper, we examine the impact of mixed tail risk on the expected stock returns from multi-dimensional perspective based on coupla method.We find that: (1) The coefficient of lower tail dependence(LTD) can capture market crashes,we can use LTD as an warning indicator for market crashes. Stocks traded on Shenzhen Main Board with strong LTD have higher future returns than that with weaker LTD, but this conclusion does not apply to the stocks traded on Small and Mid Enterprise board(SME board) and Growth Enterprise market(GEM). (2) In the period of financial crisis, the positive impact of stock mixed tail risk on stock expected return will be significantly enhanced.High circulation market capitalization and high turnover rate can reduce this impact. (3) Non-tradable Share Reform increases the liquidity of stocks,reducing the risk premium of mixed tail risk.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129015746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976173
Fatemeh Kakavandi, R. D. Reus, C. Gomes, Negar Heidari, A. Iosifidis, P. Larsen
Evaluating the product quality in an assembly machine is critical yet time-consuming since, in product assessment in batch manufacturing, a certain amount of products should be investigated in an invasive manner. However, continuous manufacturing ensures product quality assessment during assembly with high efficiency and traceability. This paper proposes a quality assessment method for an industrial use case. First, the data is prepared based on two indicators and expert knowledge. Then two data classification approaches (one-class classification and binary classification) are applied to evaluate the products’ quality by analysing the related data. Finally, the most efficient model is selected to predict the product labels and deviate anomalies from normal products. For the studied use case and the limited number of products, the binary classifier guarantees to detect 100% of defective products. The proposed approach can provide the engineers and operators with understandable extracted process knowledge, and can therefore be adapted to a high-speed manufacturing line where large data volume and process complexity can be problematic.
{"title":"Product Quality Control in Assembly Machine under Data Restricted Settings","authors":"Fatemeh Kakavandi, R. D. Reus, C. Gomes, Negar Heidari, A. Iosifidis, P. Larsen","doi":"10.1109/INDIN51773.2022.9976173","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976173","url":null,"abstract":"Evaluating the product quality in an assembly machine is critical yet time-consuming since, in product assessment in batch manufacturing, a certain amount of products should be investigated in an invasive manner. However, continuous manufacturing ensures product quality assessment during assembly with high efficiency and traceability. This paper proposes a quality assessment method for an industrial use case. First, the data is prepared based on two indicators and expert knowledge. Then two data classification approaches (one-class classification and binary classification) are applied to evaluate the products’ quality by analysing the related data. Finally, the most efficient model is selected to predict the product labels and deviate anomalies from normal products. For the studied use case and the limited number of products, the binary classifier guarantees to detect 100% of defective products. The proposed approach can provide the engineers and operators with understandable extracted process knowledge, and can therefore be adapted to a high-speed manufacturing line where large data volume and process complexity can be problematic.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115953676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976104
Eric Chiquito, Ulf Bodin, K. Synnes
In this paper, we define a multi-attribute auctioning system for the circular economy and the trade of products, components and materials subject to recycling. The increasing popularity of auctioning systems for buying and selling goods has led to the adaptation of them to diverse and particular scenarios, many of which require support for attributes like delivery time, quality, etc. Such attributes allow for more explicit and precise negotiations than traditional auctioning systems where only price is taken into account. The circular economy concept replaces end-of-life with the reuse of various goods, aiming to keep as much value as possible of any asset. By allowing users to adjust attributes in multi-step negotiations according to their economic and ecological needs, better deals can be achieved. We address this potential with our multi-attribute, and multi-step auctioning system. The system is based on transparency and fairness principles, and addresses requirements for flexibility in what attributes can be used, and the need for a semi-transparent auctioning procedure. We present a winner determination approach based on scoring protocol based on weights for different input attributes. Our auctioning system uses a signature chain data structure to provide transaction traceability. We demonstrate using a generic example that the proposed system supports simple and flexible multi-attribute auctions.
{"title":"A multi-attribute auctioning system for the circular economy with Ricardian contracts","authors":"Eric Chiquito, Ulf Bodin, K. Synnes","doi":"10.1109/INDIN51773.2022.9976104","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976104","url":null,"abstract":"In this paper, we define a multi-attribute auctioning system for the circular economy and the trade of products, components and materials subject to recycling. The increasing popularity of auctioning systems for buying and selling goods has led to the adaptation of them to diverse and particular scenarios, many of which require support for attributes like delivery time, quality, etc. Such attributes allow for more explicit and precise negotiations than traditional auctioning systems where only price is taken into account. The circular economy concept replaces end-of-life with the reuse of various goods, aiming to keep as much value as possible of any asset. By allowing users to adjust attributes in multi-step negotiations according to their economic and ecological needs, better deals can be achieved. We address this potential with our multi-attribute, and multi-step auctioning system. The system is based on transparency and fairness principles, and addresses requirements for flexibility in what attributes can be used, and the need for a semi-transparent auctioning procedure. We present a winner determination approach based on scoring protocol based on weights for different input attributes. Our auctioning system uses a signature chain data structure to provide transaction traceability. We demonstrate using a generic example that the proposed system supports simple and flexible multi-attribute auctions.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123841927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976084
J. Palmeira, Gustavo Coelho, A. Carvalho, P. Carvalhal, Paulo Cardoso
Despite the Industry 4.0, most of the production lines today are what is sometimes called "legacy", and cannot be replaced overnight by Industry 4.0 versions and thus still have to be maintained for quite some time. In this paper, we describe the architecture and implementation of a logical connector that enables the migration (also known as °to retrofit") of legacy production lines into an Industry 4.0 ecosystem, with the production lines remaining almost unchanged. To do that, four main challenges had to be addressed, namely: the data accessibility challenge, the data interoperability challenge, the machine variability challenge, and the resource usage challenge. In the end, the logical connector presented in this paper has shown to enable the migration of legacy production lines into an Industry 4.0 ecosystem and thus to reap some of the benefits promised by Industry 4.0.
{"title":"Migrating legacy production lines into an Industry 4.0 ecosystem","authors":"J. Palmeira, Gustavo Coelho, A. Carvalho, P. Carvalhal, Paulo Cardoso","doi":"10.1109/INDIN51773.2022.9976084","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976084","url":null,"abstract":"Despite the Industry 4.0, most of the production lines today are what is sometimes called \"legacy\", and cannot be replaced overnight by Industry 4.0 versions and thus still have to be maintained for quite some time. In this paper, we describe the architecture and implementation of a logical connector that enables the migration (also known as °to retrofit\") of legacy production lines into an Industry 4.0 ecosystem, with the production lines remaining almost unchanged. To do that, four main challenges had to be addressed, namely: the data accessibility challenge, the data interoperability challenge, the machine variability challenge, and the resource usage challenge. In the end, the logical connector presented in this paper has shown to enable the migration of legacy production lines into an Industry 4.0 ecosystem and thus to reap some of the benefits promised by Industry 4.0.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121257435","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976168
Nikos Piperigkos, A. Lalos, K. Berberidis
Cooperative Intelligent Transportation Systems envision the integration of cooperative intelligence as a key operational part of autonomous driving. In this way, a fleet or swarm of Connected and Automated Vehicles collectively coordinates its driving actions in order to maximize its performance. To realize this ambition, vehicles need to be fully location-aware of their surrounding environment, through distributed AI intelligence. Motivated by this requirement, we develop in this paper a distributed cooperative awareness scheme which performs multi-modal fusion of heterogeneous sensor sources along with V2V communication information, using graph Laplacian matrix and Least-Mean-Squares algorithm. The intuition behind our approach is that neighboring vehicles are interested in estimating common positions of other vehicles. We build upon our previous work on global awareness though local information diffusion, and prove that the proposed distributed framework is able to address highly efficient the case of lacking any information about other networked vehicles. More specifically, our approach achieves high enough convergence speed as well as location accuracy. The evaluation study has been performed in CARLA autonomous driving simulator and verifies the proposed method’s benefits over other related solutions.
{"title":"Robustifying cooperative awareness in autonomous vehicles through local information diffusion","authors":"Nikos Piperigkos, A. Lalos, K. Berberidis","doi":"10.1109/INDIN51773.2022.9976168","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976168","url":null,"abstract":"Cooperative Intelligent Transportation Systems envision the integration of cooperative intelligence as a key operational part of autonomous driving. In this way, a fleet or swarm of Connected and Automated Vehicles collectively coordinates its driving actions in order to maximize its performance. To realize this ambition, vehicles need to be fully location-aware of their surrounding environment, through distributed AI intelligence. Motivated by this requirement, we develop in this paper a distributed cooperative awareness scheme which performs multi-modal fusion of heterogeneous sensor sources along with V2V communication information, using graph Laplacian matrix and Least-Mean-Squares algorithm. The intuition behind our approach is that neighboring vehicles are interested in estimating common positions of other vehicles. We build upon our previous work on global awareness though local information diffusion, and prove that the proposed distributed framework is able to address highly efficient the case of lacking any information about other networked vehicles. More specifically, our approach achieves high enough convergence speed as well as location accuracy. The evaluation study has been performed in CARLA autonomous driving simulator and verifies the proposed method’s benefits over other related solutions.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127773554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976171
Wei-Yi Yang, Ya-Shu Chen, Jinqi Xiao
Resistive random-access memory (ReRAM) has been explored to be a promising solution to accelerate the inference of deep neural networks at the embedded systems by performing computations in memory. To reduce the latency of the neural network, all the pre-trained weights are pre-programmed in ReRAM cells as device resistance for the inference phase. However, the system utilization is decreased by the data dependency of the deployed neural networks and results in low energy efficiency. In this work, we propose a Lazy Engine for providing high utilization and energy-efficient ReRAM-based accelerators. Instead of avoiding idle time by applying ReRAM crossbar duplication, Lazy Engine delays the start time of the vector-matrix multiplication operations, with run-time programming overhead consideration, to reclaim idle time for energy efficiency while improving resource utilization. The experimental results show that Lazy Engine achieves up to 77% and 96% improvement in resource utilization and energy saving compared to state-of-the-art ReRAM-based accelerators.
{"title":"A Lazy Engine for High-utilization and Energy-efficient ReRAM-based Neural Network Accelerator","authors":"Wei-Yi Yang, Ya-Shu Chen, Jinqi Xiao","doi":"10.1109/INDIN51773.2022.9976171","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976171","url":null,"abstract":"Resistive random-access memory (ReRAM) has been explored to be a promising solution to accelerate the inference of deep neural networks at the embedded systems by performing computations in memory. To reduce the latency of the neural network, all the pre-trained weights are pre-programmed in ReRAM cells as device resistance for the inference phase. However, the system utilization is decreased by the data dependency of the deployed neural networks and results in low energy efficiency. In this work, we propose a Lazy Engine for providing high utilization and energy-efficient ReRAM-based accelerators. Instead of avoiding idle time by applying ReRAM crossbar duplication, Lazy Engine delays the start time of the vector-matrix multiplication operations, with run-time programming overhead consideration, to reclaim idle time for energy efficiency while improving resource utilization. The experimental results show that Lazy Engine achieves up to 77% and 96% improvement in resource utilization and energy saving compared to state-of-the-art ReRAM-based accelerators.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132342946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976072
ZhaoZhou Cai, Cong Peng, Bingyun Yang, Xiaoyue Liu
Vision-based vibration measurement technology has received extensive attention due to its advantages of non-contact, high spatial resolution, and no-load effect. However, with the complexity of measurement objects and measurement tasks, the existing visual measurement technology is gradually showing greater limitations. Specifically, due to the uncertainty of actual working conditions, not all pixels in the field of view can measure vibration. Therefore, the selection of measurement points needs to rely on prior structural information and artificial experience. Frequent manual point selection tests bring a lot of resource consumption, which greatly reduces the automation degree of visual vibration measurement. This paper focuses on an intelligent area localization method for vibration measurement of rotating machine vision and designs a deep learning-based vibration measurement area localization framework to directly feedback all reliable measurement pixels from image data, which is called the VMAL framework. Firstly, the sub-pixel physical feature information associated with vibration in the data is analyzed through an unsupervised image decomposition network, and then a regularized regional localization network is used to cluster and output reliable regional pixels. Experimental results on a medium-sized single-span rotor platform verify the effectiveness of the proposed method.
{"title":"An Intelligent Area Localization Framework for Rotating Machine Vision Vibration Measurement","authors":"ZhaoZhou Cai, Cong Peng, Bingyun Yang, Xiaoyue Liu","doi":"10.1109/INDIN51773.2022.9976072","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976072","url":null,"abstract":"Vision-based vibration measurement technology has received extensive attention due to its advantages of non-contact, high spatial resolution, and no-load effect. However, with the complexity of measurement objects and measurement tasks, the existing visual measurement technology is gradually showing greater limitations. Specifically, due to the uncertainty of actual working conditions, not all pixels in the field of view can measure vibration. Therefore, the selection of measurement points needs to rely on prior structural information and artificial experience. Frequent manual point selection tests bring a lot of resource consumption, which greatly reduces the automation degree of visual vibration measurement. This paper focuses on an intelligent area localization method for vibration measurement of rotating machine vision and designs a deep learning-based vibration measurement area localization framework to directly feedback all reliable measurement pixels from image data, which is called the VMAL framework. Firstly, the sub-pixel physical feature information associated with vibration in the data is analyzed through an unsupervised image decomposition network, and then a regularized regional localization network is used to cluster and output reliable regional pixels. Experimental results on a medium-sized single-span rotor platform verify the effectiveness of the proposed method.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115631899","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976182
Muhammed Zemzemoglu, M. Unel
In this paper, an in-situ defect detection system is proposed for automated fiber placement (AFP) process monitoring. To acquire meaningful data about the laid-up tows, the design, manufacturing and integration of a flexible three degrees of freedom vision system to the AFP machine is proposed. An image segmentation algorithm is developed to locate and isolate defects in input images. The proposed algorithm utilizes Gabor filters to extract the desired texture features which is followed by an adaptive thresholding. Successful results with four of the main defect classes namely, foreign bodies, wrinkles, gaps and bridging, were obtained. This monitoring system can reduce time-consuming and expensive efforts of manual quality inspection and will significantly increase AFP process reliability.
{"title":"Design and Implementation of a Vision Based In-Situ Defect Detection System of Automated Fiber Placement Process","authors":"Muhammed Zemzemoglu, M. Unel","doi":"10.1109/INDIN51773.2022.9976182","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976182","url":null,"abstract":"In this paper, an in-situ defect detection system is proposed for automated fiber placement (AFP) process monitoring. To acquire meaningful data about the laid-up tows, the design, manufacturing and integration of a flexible three degrees of freedom vision system to the AFP machine is proposed. An image segmentation algorithm is developed to locate and isolate defects in input images. The proposed algorithm utilizes Gabor filters to extract the desired texture features which is followed by an adaptive thresholding. Successful results with four of the main defect classes namely, foreign bodies, wrinkles, gaps and bridging, were obtained. This monitoring system can reduce time-consuming and expensive efforts of manual quality inspection and will significantly increase AFP process reliability.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128720872","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976137
Zhipeng Yuan, Shunbao Li, Po Yang, Yang Li
With the increasing demand for cost-effective crop pest management solutions, how to achieve effective and efficient automatic pest detection has become the primary research problem. Traditional object detection methods that rely on the quality of handcrafted feature selection are hardly used in pest detection due to the difficulty of designing the features of multiple types of pests. The application of deep learning which presents outstanding performances in object detection tasks faces the following challenges in the field of pest detection. First, the detection difficulties caused by tiny-size pests and protective colouration limit the accuracy of detection. Second, pest detection requires the employment of experts to obtain the annotation of pests for training models, which is costly. Finally, the ability to run on lightweight devices is required due to the limitations of the field environment on networks and equipment. To solve these problems, this paper focuses on a lightweight tiny object detection model, training on limited supervised samples through different data augmentation methods. Different components of object detection models and data augmentation methods are analysed in different sizes of training datasets. Finally, a method based on the Yolo detection model is proposed for pest detection. This pest detection model is evaluated on a real-world aphids data set containing 6k objects. Five sets of data augmentation methods are used on seven sizes of training data sets for analysis. Then the structure of the detection neck of the Yolo model is analysed. Our experimental results show that 54.35% mAP can be achieved by the PAN module and removing the Mosaic data augmentation method for tiny object detection with one hundred samples.
{"title":"Lightweight Object Detection Model with Data Augmentation for Tiny Pest Detection","authors":"Zhipeng Yuan, Shunbao Li, Po Yang, Yang Li","doi":"10.1109/INDIN51773.2022.9976137","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976137","url":null,"abstract":"With the increasing demand for cost-effective crop pest management solutions, how to achieve effective and efficient automatic pest detection has become the primary research problem. Traditional object detection methods that rely on the quality of handcrafted feature selection are hardly used in pest detection due to the difficulty of designing the features of multiple types of pests. The application of deep learning which presents outstanding performances in object detection tasks faces the following challenges in the field of pest detection. First, the detection difficulties caused by tiny-size pests and protective colouration limit the accuracy of detection. Second, pest detection requires the employment of experts to obtain the annotation of pests for training models, which is costly. Finally, the ability to run on lightweight devices is required due to the limitations of the field environment on networks and equipment. To solve these problems, this paper focuses on a lightweight tiny object detection model, training on limited supervised samples through different data augmentation methods. Different components of object detection models and data augmentation methods are analysed in different sizes of training datasets. Finally, a method based on the Yolo detection model is proposed for pest detection. This pest detection model is evaluated on a real-world aphids data set containing 6k objects. Five sets of data augmentation methods are used on seven sizes of training data sets for analysis. Then the structure of the detection neck of the Yolo model is analysed. Our experimental results show that 54.35% mAP can be achieved by the PAN module and removing the Mosaic data augmentation method for tiny object detection with one hundred samples.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127163683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：