文章摘要

文章信息/Info

Title:: Study on edge detection algorithm of urine test strip combined with Sobel operator and morphology

Author(s):: LI Yiran; PANG Chunying; CHANG Zhiqiang; School of Life Science and Technology,Changchun University of Science and Technology,Changchun 130022,China

Keywords:: Urine dipstick; Sobel operator; Mathematical morphology; Edge detection; Image processing

摘要:: 提出一种基于Sobel算子和数学形态学相结合的改进算法，用于尿液试纸条图像的边缘检测。首先采用Sobel算子对预处理后的尿液试纸条进行边缘检测，得到粗略的边缘图像；然后采用双结构多尺度形态学算子精准检测。根据图像特点，使用形态学算子时，将膨胀运算和开运算进行加权求和，有效地克服了图像的双边缘现象；最后使用顶帽底帽联合运算以增强图像边缘。实验表明，本研究提出的改进算法信噪比可达到8.6471，边缘连续性指标可达到0.9212，实现了尿液试纸条图像边缘的准确定位和清晰成像。

Abstract:: An improved algorithm based on Sobel operator and mathematical morphology was proposed to detect the edge of urine dipstick image. Firstly, the Sobel operator was used to detect the edge of the urine test dipstick, and the rough edge image was obtained. Then, the double-structure multi-scale morphological operator was used for accurate detection. According to the characteristics of the image,we used morphological operators to sum up the expansion operation and the open operation,which effectively overcomed the double edge phenomenon of the image. Finally, the top hat and bottom cap joint operation were used to enhance the edge of the image.The experimental results show that the signal-to-noise ratio of the improved algorithm can reach 8.6471, and the edge continuity index can reach 0.9212, which realizes the accurate location and clear imaging of the edge of the urine dipstick.