Identification method and system based on calcium oxalate crystal

Abstract

The invention provides an identification method and system based on calcium oxalate crystal. The method comprises the following steps: firstly, acquiring a urinary sediment image; secondly, carrying out binarization segmentation on the urinary sediment image by adopting a local adaptive threshold binarization method to obtain a binary segmentation image; determining a mark image according to the binary segmentation image; determining the number of straight line segments of the edge contour of each cell region in the marked image by adopting Hough transform; and finally, identifying calcium oxalate crystals according to the number of straight line segments of the edge contour. Factors such as uneven illumination during urinary sediment image shooting are considered. Therefore, firstly, a local adaptive threshold binarization method is adopted to carry out binarization segmentation on the urinary sediment image. Then, the number of straight line segments of the edge contour of each cellregion is determined by adopting Hough transform, and finally, the calcium oxalate crystals is identified according to the number of the straight line segments of the edge contour, so that the cell regions are accurately separated, and the accuracy of identifying the calcium oxalate crystals is further improved.

Description

technical field [0001] The invention relates to the technical field of automatic identification of calcium oxalate crystals, in particular to an identification method and system based on calcium oxalate crystals. Background technique [0002] Urinary sediment inspection is to identify, classify and count the formed elements in urine, such as red blood cells, white blood cells, epithelial cells, casts, crystals, mucus filaments, etc. Doctors conduct urinary system diagnosis according to the classification and counting results obtained by the automatic urine sediment detection equipment. Auxiliary diagnosis of diseases, thus how to accurately and automatically identify calcium oxalate crystals has become an urgent technical problem in this field. [0003] In a more efficient algorithm for automatic identification of urine sediment, a fast template matching algorithm was proposed to identify calcium oxalate crystals. The specific method is based on the fact that the interior o...

AI Technical Summary

Problems solved by technology

However, when the urine sediment image is actually taken, due to reasons such as uneven illumination, after the obtained urine sediment image is binarized, the star-shaped area in the binary image of the crystallized area cannot be extracted or the shape of the extracted star-shaped area is not star-shaped In some cases, calcium oxalate crystals cannot be accurately found by using the template matching method; at the same time, for the crystals bonded to other cells, the template matching method cannot accurately find the crystals

[0004] In the automatic identification of formed elements in microscopic cell images, it is also mentioned that the use of morphological features such as the rectangularity and narrow length of the cell area is used to automatically detect calcium oxalate crystals, but for those that are bonded to other cells or calcium oxalate crystals Calcium oxalate crystals cannot be accurately identified

Images