Recognition System For Livestock Helminth Eggs Using Invariant Moment

In this information and digital age,the traditional way to inspect of helminth eggs in feces can not meet the needs of veterinary work at present.Recognition system of livestock helminth eggs is especially important.The country has a large number of studies about human helminth eggs,However, there is little research in the field of livestock parasites. Because parasite eggs in domestic animals are considered various types,different shapes, irregular structure,which makes the image preprocessing and recognition more difficult.Firstly, we select 16 kinds of helminth eggs for image acquisition and profiling, then the information of image and words are integrated.Using Access 2003 to establish original image of database 16 kinds of helminth eggs,including 1931 original images.The database is simple and practical,building the foundation for the identification of eggs.Secondly, according to the image database,we start with a image preprocessing. Because there are all kinds of eggs, several background colors, it is difficult to find a common preprocessing method. Therefore, we choose iterative threshold segmentation and morphologic operations of erosion and swelling to determine the borderline,then according to the rectangle of image for segmentation, finally storage it.The traditional Pseudo-Jacobi(p=4,q=3)-Fourier moments(PJFM’s) is defined in the polar coordinate system. Firstly, an improved algorithm of Pseudo-Jacobi-Fourier moments, which calculated directly in the Cartesian coordinate system, is proposed in this paper. The traditional algorithm and the improved algorithm are compared by describing and reconstructing the image of capital letter E.The results show that the improved PJFM s is better that the main information of original images could be recovered by lower order of PJFM’s than original PJFM’s. Secondly, using the improved algorithm of Pseudo-Jacobi-Fourier moments to analysis invariance, the results show that invariant moments are shift-、rotation-、scale-、intensity-invariant. Then we apply this improved algorithm to classify capital letters and similar leukocyte microscopic image,a satisfactory recognition rate of 98.08% and 97.89% achieved respectively. Finally,livestock helminth eggs were digitalized and recognized by improved algorithm.Finally, we attempted to realize automatic identification system of the livestock helminth eggs with the graphic user interface (GUI) of Matlab software, linking image database,image segmentation, image recognition.With plenty of test, the operation and recognition results show that the system is simple,stable and reliable.