A Novel Electronic Chip Detection Method Using Deep Neural Networks
Electronic chip detection is widely used in electronic industries. However, most existing detection methods cannot handle chip images with multiple classes of chips or complex backgrounds, which are common in real applications. To address these problems, a novel chip detection method that combines a...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Machines |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1702/10/5/361 |
| _version_ | 1797498396865462272 |
|---|---|
| author | Huiyan Zhang Hao Sun Peng Shi Luis Ismael Minchala |
| author_facet | Huiyan Zhang Hao Sun Peng Shi Luis Ismael Minchala |
| author_sort | Huiyan Zhang |
| collection | DOAJ |
| description | Electronic chip detection is widely used in electronic industries. However, most existing detection methods cannot handle chip images with multiple classes of chips or complex backgrounds, which are common in real applications. To address these problems, a novel chip detection method that combines attentional feature fusion (AFF) and cosine nonlocal attention (CNLA), is proposed, and it consists of three parts: a feature extraction module, a region proposal module, and a detection module. The feature extraction module combines an AFF-embedded CNLA module and a pyramid feature module to extract features from chip images. The detection module enhances feature maps with a region intermediate feature map by spatial attentional block, fuses multiple feature maps with a multiscale region of the fusion block of interest, and classifies and regresses objects in images with two branches of fully connected layers. Experimental results on a medium-scale dataset comprising 367 images show that our proposed method achieved <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>m</mi><mi>A</mi><msup><mi>P</mi><mrow><mn>0.5</mn></mrow></msup><mo>=</mo><mn>0.98745</mn></mrow></semantics></math></inline-formula> and outperformed the benchmark method. |
| first_indexed | 2024-03-10T03:32:52Z |
| format | Article |
| id | doaj.art-70e45e6481494e4ba2b7bcee4b1cc6ea |
| institution | Directory Open Access Journal |
| issn | 2075-1702 |
| language | English |
| last_indexed | 2024-03-10T03:32:52Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Machines |
| spelling | doaj.art-70e45e6481494e4ba2b7bcee4b1cc6ea2023-11-23T11:52:49ZengMDPI AGMachines2075-17022022-05-0110536110.3390/machines10050361A Novel Electronic Chip Detection Method Using Deep Neural NetworksHuiyan Zhang0Hao Sun1Peng Shi2Luis Ismael Minchala3National Research Base of Intelligent Manufacturing Service, Chongqing Technology and Business University, Chongqing 400067, ChinaBio-Computing Research Center, Harbin Institute of Technology, Shenzhen 518055, ChinaSchool of Electrical and Electronic Engineering, University of Adelaide, Adelaide, SA 5005, AustraliaDepartment of Electrical Electronics and Telecommunications Engineering, University of Cuenca, Cuenca 010105, EcuadorElectronic chip detection is widely used in electronic industries. However, most existing detection methods cannot handle chip images with multiple classes of chips or complex backgrounds, which are common in real applications. To address these problems, a novel chip detection method that combines attentional feature fusion (AFF) and cosine nonlocal attention (CNLA), is proposed, and it consists of three parts: a feature extraction module, a region proposal module, and a detection module. The feature extraction module combines an AFF-embedded CNLA module and a pyramid feature module to extract features from chip images. The detection module enhances feature maps with a region intermediate feature map by spatial attentional block, fuses multiple feature maps with a multiscale region of the fusion block of interest, and classifies and regresses objects in images with two branches of fully connected layers. Experimental results on a medium-scale dataset comprising 367 images show that our proposed method achieved <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>m</mi><mi>A</mi><msup><mi>P</mi><mrow><mn>0.5</mn></mrow></msup><mo>=</mo><mn>0.98745</mn></mrow></semantics></math></inline-formula> and outperformed the benchmark method.https://www.mdpi.com/2075-1702/10/5/361electronic chip detectiondeep learningfeature pyramid network |
| spellingShingle | Huiyan Zhang Hao Sun Peng Shi Luis Ismael Minchala A Novel Electronic Chip Detection Method Using Deep Neural Networks Machines electronic chip detection deep learning feature pyramid network |
| title | A Novel Electronic Chip Detection Method Using Deep Neural Networks |
| title_full | A Novel Electronic Chip Detection Method Using Deep Neural Networks |
| title_fullStr | A Novel Electronic Chip Detection Method Using Deep Neural Networks |
| title_full_unstemmed | A Novel Electronic Chip Detection Method Using Deep Neural Networks |
| title_short | A Novel Electronic Chip Detection Method Using Deep Neural Networks |
| title_sort | novel electronic chip detection method using deep neural networks |
| topic | electronic chip detection deep learning feature pyramid network |
| url | https://www.mdpi.com/2075-1702/10/5/361 |
| work_keys_str_mv | AT huiyanzhang anovelelectronicchipdetectionmethodusingdeepneuralnetworks AT haosun anovelelectronicchipdetectionmethodusingdeepneuralnetworks AT pengshi anovelelectronicchipdetectionmethodusingdeepneuralnetworks AT luisismaelminchala anovelelectronicchipdetectionmethodusingdeepneuralnetworks AT huiyanzhang novelelectronicchipdetectionmethodusingdeepneuralnetworks AT haosun novelelectronicchipdetectionmethodusingdeepneuralnetworks AT pengshi novelelectronicchipdetectionmethodusingdeepneuralnetworks AT luisismaelminchala novelelectronicchipdetectionmethodusingdeepneuralnetworks |