Cell And Jinbiao Detection Algorithm Based On Image Processing And The Project Implementation

In the era of progress, along with computer science and technology in the continuous development, the demand for image information have gradually increased in their daily lives,digital image processing technology was developed in recent years has been rapid. Due to digital image processing technology for its large amount of information, processing and transport convenience, a wide range of applications and a series of advantages in space exploration, biomedicine, industrial and agricultural production, military, public security,office automation, and other fields has been widely used, showing broad application prospects,The article is in the image processing application in biomedical detection.This thesis is based on a medical testing platform, the system can be realized using digital image processing technology, real-time processing of blood samples images to a computer with an optical microscope connected to sampling, image processing, the statistics of the image in red white number, detected blood negative or positive and detect other components. The QT embedded in the Visual Studio 2010 environment to achieve the main work in the following areas:First, the development and application of domestic and foreign cell recognition technology to do a comprehensive research on the digital image processing techniques denoising, edge detection, image smoothing, image binarization, image morphological changes in the filling hole corrosion and images were systematically studied the pros and cons of each method are analyzed comparison.Second, according to the system design goals and functional requirements, the design of the overall architecture and workflow systems, introduces the function of each module.Third, the effect of system functions to analyze the effectiveness of the verification system function and practicality. Respectively, image smoothing, image binarization, image morphological transformation effects and recognition accuracy of the final cell were analyzedand evaluated.On the basis of a detailed analysis of the system requirements on the system design and implementation of applied clinical research and patient diagnostic reference case. Designed system used to identify statistics can improve efficiency, the red white image test system cell recognition rate remained at around 90%, and gold assay color recognition rate of between85% to 90%, indicating that the system has higher recognition accuracy statistics, combined with the system is reliable, ease of use, scalability, relevance and practical features, used in clinical diagnosis has a certain significance.