Research On Image Threshold Segmentation Arithmetic Based On OTSU

Image segmentation is the process and related methods to segment an image into different sub-images with different characters and derive some interested objects, it is a key step of image process, the quality of segmentation will directly affect the follow-on work of image process. So the study on the image segmentation is always one of the key points in the image segmentation technique study, many image segmentation algorithms are advanced, and it is widely used in many fields.Threshold segmentation method is common use in the image segmentation. It distinguishes the image into object and background by picking threshold, so the threshold picking is the key technique.First of all, this paper introduces the main image segmentation algorithm and the commonly used threshold segmentation method, and analyzes the advantages and disadvantages by studying a large number of references.Secondly, this paper introduces the one-dimensional and two-dimensional Otsu (the Otsu) method in detail, and analysis the two-dimensional Otsu method has operation for a long time, and bad real-time. On this basis, this paper proposes the change of two-dimensional histogram to improve the computing efficiency and real-time of two-dimensional Otsu method, and experiments verify.Finally, this paper introduces the improvemented of three-dimensional Otsu Threshold in detail. The segmentation effect of two-dimensional Otsu thresholding has dropped constantly with the increase of noise, hence the lowering of SNR. Now, the average vaule in the neighborhood has been introduced as the third characteristic and the histogram of three-dimensional gray level is structured, based on which the three-dimensional Otsu thresholding has been proposed, and the better result of segmenting has been obtained for the image with low contrast ratio and low SNR, and it has been dropped the time complexity by recurrence method. Although the reduction of operation time, it can not meet the requirements of real-time, therefore, this paper blend the immune genetic algorithm into three-dimensional Otsu thresholding, and use the recursive method to calculate the fitness value. This paper not only get the optimal threshold by using the global search feature of IGA, but also improve the operation time by using the characteristics of the fast convergence, and improve the real-time of three-dimensional Otsu thresholding, and experimental validation.