Precise Detection for Dense PCB Components Based on Modified YOLOv8
Effective detection of dense printed circuit board (PCB) components contributes to the optimization of automatic flow of production. In addition, PCB component recognition is also the essential prerequisite for early defect detection. Current PCB component detection approaches are not adept in both...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2023-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10287971/ |
_version_ | 1797649803246567424 |
---|---|
author | Qin Ling Nor Ashidi Mat Isa Mohd Shahrimie Mohd Asaari |
author_facet | Qin Ling Nor Ashidi Mat Isa Mohd Shahrimie Mohd Asaari |
author_sort | Qin Ling |
collection | DOAJ |
description | Effective detection of dense printed circuit board (PCB) components contributes to the optimization of automatic flow of production. In addition, PCB component recognition is also the essential prerequisite for early defect detection. Current PCB component detection approaches are not adept in both rapid and precise detection. YOLOv8 models have exhibited effective performances for detecting common objects, such as person, car, chair, dog etc. However, it is still tricky for YOLOv8 models to inspect dense and disparate PCB components precisely. Thus, a novel convolution neural network (CNN) model is proposed for dense PCB component detection by introducing several modifications onto YOLOv8. First, creative C2Focal module is designed as the core element of the backbone, combining both fine-grained local and coarse-grained global features concurrently. Then, the lightweight Ghost convolutions are inserted to effectively reduce the computation cost, meanwhile maintaining the detection performance. Finally, a new bounding box regression loss that is Sig-IoU loss, is proposed to facilitate the prediction regression and promote the positioning accuracy. The experiments on our PCB component dataset demonstrate that our proposed model performs the highest mean average precisions of 87.7% (mAP@0.5) and 75.3% (mAP@0.5:0.95) respectively, exceeding other state-of-the-arts. Besides, the detection speed hits 110 frames per second using RTX A4000, which is potential to realize the real-time inspection. |
first_indexed | 2024-03-11T15:51:16Z |
format | Article |
id | doaj.art-bcaebdd16f974b9985b89493ce7e6688 |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-03-11T15:51:16Z |
publishDate | 2023-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-bcaebdd16f974b9985b89493ce7e66882023-10-25T23:01:12ZengIEEEIEEE Access2169-35362023-01-011111654511656010.1109/ACCESS.2023.332588510287971Precise Detection for Dense PCB Components Based on Modified YOLOv8Qin Ling0https://orcid.org/0000-0003-0957-4804Nor Ashidi Mat Isa1https://orcid.org/0000-0002-2675-4914Mohd Shahrimie Mohd Asaari2School of Electrical and Electronic Engineering, Engineering Campus, Universiti Sains Malaysia, Nibong Tebal, Pulau Pinang, MalaysiaSchool of Electrical and Electronic Engineering, Engineering Campus, Universiti Sains Malaysia, Nibong Tebal, Pulau Pinang, MalaysiaSchool of Electrical and Electronic Engineering, Engineering Campus, Universiti Sains Malaysia, Nibong Tebal, Pulau Pinang, MalaysiaEffective detection of dense printed circuit board (PCB) components contributes to the optimization of automatic flow of production. In addition, PCB component recognition is also the essential prerequisite for early defect detection. Current PCB component detection approaches are not adept in both rapid and precise detection. YOLOv8 models have exhibited effective performances for detecting common objects, such as person, car, chair, dog etc. However, it is still tricky for YOLOv8 models to inspect dense and disparate PCB components precisely. Thus, a novel convolution neural network (CNN) model is proposed for dense PCB component detection by introducing several modifications onto YOLOv8. First, creative C2Focal module is designed as the core element of the backbone, combining both fine-grained local and coarse-grained global features concurrently. Then, the lightweight Ghost convolutions are inserted to effectively reduce the computation cost, meanwhile maintaining the detection performance. Finally, a new bounding box regression loss that is Sig-IoU loss, is proposed to facilitate the prediction regression and promote the positioning accuracy. The experiments on our PCB component dataset demonstrate that our proposed model performs the highest mean average precisions of 87.7% (mAP@0.5) and 75.3% (mAP@0.5:0.95) respectively, exceeding other state-of-the-arts. Besides, the detection speed hits 110 frames per second using RTX A4000, which is potential to realize the real-time inspection.https://ieeexplore.ieee.org/document/10287971/PCB component detectionhigh-precisionlightweightSig-IoU lossghost convolution |
spellingShingle | Qin Ling Nor Ashidi Mat Isa Mohd Shahrimie Mohd Asaari Precise Detection for Dense PCB Components Based on Modified YOLOv8 IEEE Access PCB component detection high-precision lightweight Sig-IoU loss ghost convolution |
title | Precise Detection for Dense PCB Components Based on Modified YOLOv8 |
title_full | Precise Detection for Dense PCB Components Based on Modified YOLOv8 |
title_fullStr | Precise Detection for Dense PCB Components Based on Modified YOLOv8 |
title_full_unstemmed | Precise Detection for Dense PCB Components Based on Modified YOLOv8 |
title_short | Precise Detection for Dense PCB Components Based on Modified YOLOv8 |
title_sort | precise detection for dense pcb components based on modified yolov8 |
topic | PCB component detection high-precision lightweight Sig-IoU loss ghost convolution |
url | https://ieeexplore.ieee.org/document/10287971/ |
work_keys_str_mv | AT qinling precisedetectionfordensepcbcomponentsbasedonmodifiedyolov8 AT norashidimatisa precisedetectionfordensepcbcomponentsbasedonmodifiedyolov8 AT mohdshahrimiemohdasaari precisedetectionfordensepcbcomponentsbasedonmodifiedyolov8 |