Engineering Research On The Features Extraction And Recognition Of Image Of Human Helminth Eggs

This paper discusses the relevant problem about how to develop a micro-image recognition system of human helminth eggs.In theory, every problem of involving morphological analysis and examination can be solved with digital image processing technology. Because there are irregular form structures, uncertain space positions and impure samples of the human helminth eggs, digital application and research is underdeveloped in parasitology. The recognition and analysis of human helminth eggs is unable to be performed by means of automatic instrument as blood cell. As a result, they have been observed and distinguished under the microscope through the naked eye for a long time. The artificial recognition lacks the objectivity and accuracy, and the discriminating capability often depends on the experience and spirits of the recognizer, and it is inconvenient to save, repeat and query for the testing sample image, data and results, which can't meet the needs of the information-based development of modern hospitals.The key to realize the automatic recognition by means of computer vision and the pattern-recognition technology is to develop the research on the feature's extraction and recognition of the images of human helminth eggs. In this research area, several domestic and international similar projects have some problems generally, such as the unaccurate image characteristic chooses, the low classified precision, the slow software processing speed, the unablely automatic processing, and so on.This thesis explores the relevant contents about digital image processing technology, pattern-recognition technology and the image's features of 10 common human helminth eggs. Taking Visual C++6.0 as the system development kit, using digital image processing technology and passing the pretreatment, we can extract some features of 10 common human body helminth eggs' digital images collected by micro-camera, such as conformation, color, texture, etc. Through corresponding categorized algorithm to helminth eggs in the computer, we have structured The Image Recognition System of Common Human Helminth Eggs (OvRecognizer 1.0), and have realized tentatively the automatic image recognition of human common helminth eggs.Through the research on analysis, selecting and extracting about the color features, and texture features of helminth eggs' images and by means of a large...