Design of a System for Detecting the Number of Holes on the Surface of Filter Parts Based on Image Stitching
Year: 2025
Type: article
Abstract: 本文提出一种基于RR-SIFT图像拼接与LabVIEW孔数检测的综合系统，用于解决滤网零件孔数不确定问题。为克服相机视野限制，利用改进的RR-SIFT算法实现高精度特征匹配，将局部侧面图拼接成全景图，然后对滤网表面孔数进行检测。实验表明，改进算法匹配准确率超过94%，而传统SIFT仅约77%~80%；最终孔数检测准确率达99.5%以上，检测时间控制在1.5~3.7秒内。该系统显著提升了检测效率和精度，为滤网质量控制提供了可靠支撑，具有较高的工业应用潜力。In this paper, an integrated system based on RR-SIFT image stitching and LabVIEW hole number detection is proposed for solving the problem of uncertainty in the number of holes in a screen part. To overcome the limitation of camera field of view, the improved RR-SIFT algorithm is utilized to achieve high-precision feature matching, and the local side view is spliced into a panoramic view, and then the number of holes on the surface of the filter screen is detected. Experiments show that the matching accuracy of the improved algorithm is more than 94%, while that of the traditional SIFT is only about 77% to 80%; the final accuracy of the hole number detection is more than 99.5%, and the detection time is controlled within 1.5 to 3.7 seconds. The system significantly improves the detection efficiency and accuracy, provides reliable support for filter quality control, and has high potential for industrial application.
Author: 啸天 肖
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4411612017