Quantitative colloidal gold test paper detection method and system based on computer vision

The invention discloses a quantitative colloidal gold test paper detection method and system based on computer vision. The quantitative colloidal gold test paper detection method comprises the following steps: collecting a test paper picture by adopting a mobile phone camera, carrying out image recognition by combining a digital image processing method, extracting target characteristics, and positioning a T-line section. And taking the color signal as a region of interest (ROI) and quantifying the color signal of the region. A series of standard test paper is prepared for detection, and a regression model of a T-line developing signal-detected substance concentration is obtained, so that a sample to be detected is quantified. The method has the advantages of high precision, high efficiency, convenience, expandability, flexibility, high universality and low cost. 本发明公开了一种基于计算机视觉的定量胶体金试纸检测方法和系统，包括：采用手机摄像头采集试纸图片，结合数字图像处理方法，进行图像识别提取目标特征，定位T线区段。以此作为兴趣区域(ROI)并量化该区域显色信号。通过配制系列标品试纸检测，获取T线显色信号-被测物浓度的回归模型，从而定量待测样品。本发明的