An efficient algorithm for detecting spherical particles from differential interference microscopic images. (14th July 2017)

• Record Type: Journal Article
• Authors: Hong, Hyeyoung; Nam, Jaewook
• Is Part Of: Measurement science & technology. Volume 28:Number 8(2017:Aug.)
• Publication Date: 2017-07-14
• Subjects: image analysis -- differential interference microscope -- anisotropic conductive film -- particle detection -- spherical particle; Physical measurements -- Periodicals; Scientific apparatus and instruments -- Periodicals; Equipment and Supplies -- Periodicals; Science -- instrumentation -- Periodicals; Technology -- instrumentation -- Periodicals; Mesures physiques -- Périodiques; Physical measurements; Scientific apparatus and instruments; Periodicals; 502.87
• DOI: 10.1088/1361-6501/aa7442
• Languages: English
• ISSNs: 0957-0233

Abstract

Abstract: Particulate films have become critical parts in electronics industries because they can provide the required optical and electrical properties. Microstructures, which consist of various particles inside the films, strongly influence the performance of devices. Therefore, the visualization and detection of particles can be an indirect method of determining the device performance without destroying samples. In this study, we propose an image analysis algorithm to detect spherical particles inside a thin film visualized by a differential interference contrast microscope. Our goal is to develop a robust and efficient algorithm specifically for uniform-sized spherical particles sandwiched between two solid thin and flat plates, which are usually applied in anisotropic conductive films that provide electrical and mechanical connections between two electrodes from a chip and substrate. Our algorithm uses correlation and masking to detect particle centers and algebraic operations to reconstruct particle shapes. Despite the simple processes involved, our algorithm has yielded accurate results in particle detection compared with manual detections and synthetically generated ground-truth images.