In pursuit of cost-effective manufacturing, enterprises are increasingly adopting the practice of utilizing recycled semiconductor chips. To ensure consistent chip orientation during packaging, a circular marker on the front side is employed for pin alignment following successful functional testing. However, recycled chips often exhibit substantial surface wear, and the identification of the relatively small marker proves challenging. Moreover, the complexity of generic target detection algorithms hampers seamless deployment. Addressing these issues, this paper introduces a lightweight YOLOv8s-based network tailored for detecting markings on recycled chips, termed Van-YOLOv8. Initially, to alleviate the influence of diminutive, low-resolution markings on the precision of deep learning models, we utilize an upscaling approach for enhanced resolution. This technique relies on the Super-Resolution Generative Adversarial Network with Extended Training (SRGANext) network, facilitating the reconstruction of high-fidelity images that align with input specifications. Subsequently, we replace the original YOLOv8s model’s backbone feature extraction network with the lightweight Vanilla Network (VanillaNet), simplifying the branch structure to reduce network parameters. Finally, a Hybrid Attention Mechanism (HAM) is implemented to capture essential details from input images, improving feature representation while concurrently expediting model inference speed. Experimental results demonstrate that the Van-YOLOv8 network outperforms the original YOLOv8s on a recycled chip dataset in various aspects. Significantly, it demonstrates superiority in parameter count, computational intricacy, precision in identifying targets, and speed when compared to certain prevalent algorithms in the current landscape. The proposed approach proves promising for real-time detection of recycled chips in practical factory settings.

To address the challenge of automatic recognition of electronic components on an assembly line, an improved YOLOv5 was used to implement instance segmentation of four categories of electronic components. Firstly, multi-channel histogram equalization was used for image preprocessing. Then, the YOLOv5 was improved: Segmentation head was added; Sequeeze-and-excitation net(SE-Net) channel attention module was embedded to enhance the feature extraction capability and to compress the useless information without increasing the model complexity; GhostNet was used to make the model lightweight; and BiFPN was used to enhance model feature fusion capability. Finally, experimental results showed that the mAP of the proposed method could reach 96.7% and the detection time of a single image was 45.5 ms. The results prove that proposed method has superior performance than that based on mask region-based conventional neural network(Mask RCNN) and initial YOLOv5, and has practical significance for automatic detection of electronic components.