Cybid (Poland)

ABSTRACT This paper introduces an innovative approach to enhancing the quality of electronic component assembly through real-time, inline inspection utilizing AI-powered deep learning techniques. The primary objective is to ensure that each component meets the highest manufacturing standards, specifically adhering to IPC-A-610 and IPC-J-STD-001 criteria. The methodology leverages the existing infrastructure of pick-and-place machines to capture high-resolution images of electronic components during the assembly process. These images are analyzed in real-time by advanced AI deep learning algorithms, which are designed to detect defects such as damage, corrosion, and structural irregularities in the components and their leads. This AI-driven solution shifts quality assurance from a reactive to a proactive approach. Key elements of the proposed method include the integration of AI deep learning technology, real-time defect detection, and strict adherence to industry standards. By embedding inline inspection capabilities into the electronic component assembly workflow, manufacturers can proactively identify and rectify defects during the assembly process, thereby significantly enhancing overall manufacturing quality and reliability. This proactive approach anticipates defects before they manifest, leading to fewer production disruptions, smoother production flow, and ultimately, cost savings. This paper presents various examples of defective components, illustrating different types of defects identified using AI deep learning methods. Through practical applications and results, this research provides valuable insights for optimizing electronic component assembly processes. The adoption of AI technologies in EMS elevates production efficiency and ensures unparalleled quality, positioning manufacturers to achieve higher standards in electronic manufacturing.

This paper presents an algorithm for ellipse detection on stains of blood, which is directly suited for the needs of forensic analysis. The algorithm is of the edge-analyzing type. It performs convexity detection and is able to split contours of overlapping ellipses by concave regions analysis. A filtering is applied to the fit ellipses to reduce the results only to those results necessary for blood drop trajectory tracing. Results for running time and fitting quality tests performed on real-life and artificial data are presented. The solution answers to both quality and running time expectations of the field of application.

—