The Identification And Detection Of Brans In Flour

The brans contained in the flour is an important indicator to evaluation the flour's rank, which also reflects the technique of production level. In the article, the system of "the identification and detection of brans in flour " can be used for flour production to monitor the production of flour for real time on-line. It also can be used as flour testing organizations detection apparatus to detect the samples flour.The system is based on the relevant theories of digital image processing and pattern recognition. First, the article described the structure of the system block diagram in detail, then introduced the functions of every parts of the system and how to achieve the functions. The System algorithm mainly includes image pre-processing algorithm and the recognition and measurement of brans in flour algorithm. First step of image pre-processing algorithm is convert the RGB true Color image to gray image, and then discussed the median filtering, Gaussian filtering, morphological filtering, Laplacian sharpening, etc. algorithm, and finally using the Gaussian smoothing filter combined with the Laplacian sharpening. Based on the pre-processing image, the paper discussed different algorithm to detect and measure brans in flour, including template matching method, linear discriminant method, K-means clustering method and other methods. Finally, according to the flour and brans have different gray values, dividing the image into small blocks, then calculating the gray value of each single block, and decided which is the best threshold of the image. Next step is calculating the number and size of the brans in the image. After that is to analyze the contrast of the image, obtaining useful information, displaying the data obtained on the display device.The system detected the quality of flour in three aspects : the number of brans, brans share the proportion of flour, the size of each bran. At the end of the article, there is a sum of the advantages and disadvantages of the system, and taking into account the characteristics of other online testing instruments, there gives some improvements.