Robust BGA Void Detection Using Multi Directional Scan Algorithms
Authors:
Vikas Ahuja,
Vijay Kumar Neeluru
Abstract:
The life time of electronic circuits board are impacted by the voids present in soldering balls. The quality inspection of solder balls by detecting and measuring the void is important to improve the board yield issues in electronic circuits. In general, the inspection is carried out manually, based on 2D or 3D X-ray images. For high quality inspection, it is difficult to detect and measure voids…
▽ More
The life time of electronic circuits board are impacted by the voids present in soldering balls. The quality inspection of solder balls by detecting and measuring the void is important to improve the board yield issues in electronic circuits. In general, the inspection is carried out manually, based on 2D or 3D X-ray images. For high quality inspection, it is difficult to detect and measure voids accurately with high repeatability through the manual inspection and it is time consuming process. In need of high quality and fast inspection, various approaches were proposed for void detection. But, lacks in robustness in dealing with various challenges like vias, reflections from the plating or vias, inconsistent lighting, noise, void-like artefacts, various void shapes, low resolution images and scalability to various devices. Robust BGA void detection becomes quite difficult problem, especially if the image size is very small (say, around 40x40) and with low contrast between void and the BGA background (say around 7 intensity levels on a scale of 255). In this work, we propose novel approach for void detection based on the multi directional scanning. The proposed approach is able to segment the voids for low resolution images and can be easily scaled to various electronic manufacturing products.
△ Less
Submitted 31 August, 2019;
originally announced September 2019.
Void region segmentation in ball grid array using u-net approach and synthetic data
Authors:
Vijay Kumar Neeluru,
Vikas Ahuja
Abstract:
The quality inspection of solder balls by detecting and measuring the void is important to improve the board yield issues in electronic circuits. In general, the inspection is carried out manually, based on 2D or 3D X-ray images. For high quality inspection, it is difficult to detect and measure voids accurately with high repeatability through the manual inspection and the process is time consumin…
▽ More
The quality inspection of solder balls by detecting and measuring the void is important to improve the board yield issues in electronic circuits. In general, the inspection is carried out manually, based on 2D or 3D X-ray images. For high quality inspection, it is difficult to detect and measure voids accurately with high repeatability through the manual inspection and the process is time consuming. In need of high quality and fast inspection, various approaches were proposed, but, due to the various challenges like vias, reflections from the plating or vias, inconsistent lighting, noise and void-like artifacts makes these approaches difficult to work in all these challenging conditions. In recent times, deep learning approaches are providing the outstanding accuracy in various computer vision tasks. Considering the need of high quality and fast inspection, in this paper, we applied U-Net to segment the void regions in soldering balls. As it is difficult to get the annotated dataset covering all the variations of void, we proposed an approach to generated the synthetic dataset. The proposed approach is able to segment the voids and can be easily scaled to various electronic products.
△ Less
Submitted 7 July, 2019;
originally announced July 2019.
