Efficient Faster R-CNN: Used in PCB Solder Joint Defects and Components Detection

Fan, Fangliang; Wang, Baoyu; Zhu, Gailin; Wu, Jianhua

doi:10.1109/ccet52649.2021.9544356

Cited by 18 publications

(9 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The methodology comprises three primary phases: data augmentation, model construction, and model evaluation, with the selection of YOLOv4 for further refinement based on its successful application in PCB defect detection as demonstrated by Caliskan and Gurkan [17] and Xin et al [18]. This proposed solution aims to improve YOLOv4 by substituting the original CSPDarknet-53 backbone with EfficientNet, a proven effective backbone demonstrated by Fan et al [3] for detecting PCB solder joint defects and components. EfficientNet's success extends beyond the PCB industry; for instance, a modified YOLOv4 with EfficientNet-B0 as its backbone has been utilized in apple detection, resulting in a lighter model with reduced computational complexity and enhanced performance compared to YOLOv3 and YOLOv4 [32].…”

Section: Methodsmentioning

confidence: 99%

“…Several loss functions have been introduced in component detection to improve the object detector. Examples include Generalized Intersection over Union (GIoU) [3], Gaussian Intersection of Union (GsIoU) [20], Loss Boosting (LB) [10], and modified binary crossentropy (BCE) [19].…”

Section: ) One-stage Detector: Yolomentioning

confidence: 99%

“…Moreover, a specific variant of Faster R-CNN, Inception-v2, exhibited promising performance in localizing PCB components, particularly in identifying absent resistors [25]. Innovations also encompass an enhanced Faster R-CNN version, Efficient Faster R-CNN, introduced by Fan et al [3]. This variant utilizes the EfficientNet-B7 network to accurately detect solder joint defects and PCB components while replacing the original VGG-16 backbone network.…”

Section: ) Two-stage Detectormentioning

confidence: 99%

“…Printed circuit boards serve as the core of electronic devices, housing interconnected circuits and electronic components. Traditional manual visual inspection, prone to inefficiencies and errors due to human limitations, prompts the need for non-contact automation methods, particularly in PCB assembly (PCBA) [3]. Automatic Optical Inspection systems (AOIs) have emerged as a pivotal application of machine vision to ensure PCB quality and streamline inspection processes, alleviating manual inspection challenges [4].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Enhancing EfficientNet-YOLOv4 for Integrated Circuit Detection on Printed Circuit Board (PCB)

Chi,

Wahab,

Mohamed

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Ensuring printed circuit board (PCB) quality and functionality during manufacturing demands precise automated visual inspection. Detecting integrated circuits (ICs) on PCBs poses a significant challenge due to diverse component sizes, types, and intricate board markings, complicating accurate object detection. This study addresses this challenge by proposing an enhanced EfficientNet-YOLOv4 algorithm tailored explicitly for IC detection on PCBs. Integrating EfficientNet's robust feature extraction capabilities with YOLOv4's precise object localization, the methodology employs diverse data augmentation techniques to enrich the training dataset. Extensive experiments showcase the algorithm's efficacy and robustness in handling complex PCB layouts and varying lighting conditions, outperforming existing PCB inspection models. The proposed method, EfficientNetv2-L-YOLOv4, achieves an impressive F1-score of 99.22 with an inference speed of 0.135 seconds. The proposed method also performed well compared to EfficientNet-B7-FasterRCNN and original YOLOv4; it attains an F1-score of 98.96 and an inference speed of 0.102 seconds (with a batch size of 4). These results underscore the significance of effective feature extraction networks in object detection. Beyond addressing IC detection challenges, this algorithm advances computer vision and object detection fields. Implementing EfficientNetv2-L-YOLOv4 in real manufacturing scenarios holds promise for automating component inspection, potentially eliminating human intervention. INDEX TERMSAutomated visual inspection, Feature extraction networks, Object detection, Printed circuit board (PCB).

show abstract

Section: Methodsmentioning

confidence: 99%

Section: ) One-stage Detector: Yolomentioning

confidence: 99%

Section: ) Two-stage Detectormentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Enhancing EfficientNet-YOLOv4 for Integrated Circuit Detection on Printed Circuit Board (PCB)

Chi,

Wahab,

Mohamed

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Moreover, the scarcity of unqualified samples in real assembly lines leads to an unbalanced problem, which is challenging for classification by means of classifiers. Although deep learning was an alternative solution to avoid the problem of feature extraction in previous inspection methods (Fan et al , 2021; Wu et al , 2019), a large number of qualified and unqualified samples were required to self-learn inherent features from their images, which is unrealistic in real electronic assembly lines.…”

Section: Introductionmentioning

confidence: 99%

IC solder joint inspection via adaptive statistical modeling

Xiao

Cai

et al. 2022

SSMT

View full text Add to dashboard Cite

Purpose Statistical modeling has been successfully applied to integrated circuit (IC) solder joint inspection. However, there are some inherent problems in previous statistical modeling methods. This paper aims to propose an adaptive statistical modeling method to further improve the inspection performance for IC solder joints. Design/methodology/approach First, different pixels in the IC solder joint image were modeled by different templates, each of which was composed of the hue value of the pixel and a proposed template significance factor. Then, the potential defect image was obtained by adaptive template matching and the potential defect threshold for each pixel. It was noted that the number of templates, matching distance threshold, potential defect threshold and updating rate were adaptively updated during model training. Finally, the trained statistical model was used to inspect the IC solder joints by means of defect degree. Findings Experimental results indicated that the proposed adaptive schemes greatly contributed to the inspection performance of statistical modeling. Also, the proposed inspection method achieved better performance compared with some state-of-the-art inspection methods. Originality/value The proposed method offers a promising approach for IC solder joint inspection, which establishes different numbers of templates constructed by pixel values and template significance factors for different pixels. Also, some important parameters were adaptively updated with the updating of the model, which contributed to the inspection performance of the model.

show abstract