Based on the inherent characteristics of the images of pigs using B-ultrasound machines, several methods suitable for B-ultrasound image processing have been selected from numerous image enhancement methods. According to basic theories and a large number of image enhancement experiments, algorithms that can be used to enhance images of pigs using B-ultrasound machines have been identified and implemented in B-ultrasound images.

(1) The use of histogram equalization enhancement method can enhance the required B-ultrasound images to a certain extent, but the problem of excessive computational complexity and slow speed in local histogram equalization is more prominent. Moreover, the image distortion is relatively large, not smooth enough, and there is no selectivity. It cannot find the best enhancement image by changing parameters like the other three enhancement methods. It is only suitable for simple Pig ultrasound images that do not require too much detail. So it is still necessary to continue optimizing the algorithm and reducing the computational load.

(2) The median filtering method has poor enhancement effect on B-ultrasound images with average clarity and no obvious noise points. Although it is suitable for eliminating isolated noise points, such as salt and pepper noise. This greatly limits the application of median filtering in the field of image enhancement.

Pig B-ultrasound machine

(3) Wavelet enhancement has special advantages over traditional enhancement methods. It not only simplifies the filtering operation, but also effectively filters out useless information in the original B-ultrasound image. Moreover, it is more convenient to process the image in the wavelet domain, making the details of the lesion in the processed pig B-ultrasound machine image clearer and improving the level of intuitive observation. The enhancement effect is relatively good, and the computational complexity is moderate, making it easier to implement and control. There are many methods for wavelet enhancement, and this article mainly introduces the sub-band enhancement method and the anti sharpening mask method, which are suitable for image enhancement of pigs using B-ultrasound Machines. The two wavelet enhancement methods also have their own advantages and disadvantages. The enhanced B-ultrasound image based on wavelet sub-band enhancement method has a more layered effect than the original image, but due to the limitations of the method itself, the image clarity is lower and some details are not clearly highlighted. The image processed by the anti sharpening mask method not only enhances the contour lines, but also makes the image clearer. Compared with these two methods, the effect of the circle image enhanced by the anti sharpening mask method is better. Although there is a weak ringing phenomenon in the enhanced image, it has a strong sense of hierarchy and clearer details. The enhancement effect is particularly obvious for the lesion group that needs to be highlighted. Overall, the image quality has been greatly improved.

(4) Traditional fuzzy enhancement methods (classic Pal and Ki female fuzzy enhancement algorithms) cannot quickly determine the optimal parameters. It is necessary to compare the enhanced images, and in order to improve traditional fuzzy enhancement methods, the focus is on how to achieve the selection of optimal parameters. On the basis of overcoming various shortcomings of the classic Pa algorithm, we chose a fuzzy enhancement algorithm based on grayscale transformation. However, its fuzzy enhancement effect depends more on the selection of threshold values. Therefore, this article adds an automatic threshold value selection algorithm to make this method more perfect, which is also one of the innovative points of this article. It has good smoothness and selectivity, and its computational cost is not large or time-consuming, making it a feasible image enhancement method. Moreover, with the addition of an automatic threshold selection algorithm, the two are effectively combined, enabling it to automatically select the best threshold value for enhancement of different B-ultrasound images, thereby effectively enhancing the prominent parts required for B-ultrasound images.

In summary, this article verifies through experiments that the wavelet transform based anti sharpening mask method and the improved gray level based fuzzy enhancement method can adapt to the image processing of pig B-ultrasound machines under different medical diagnostic needs, significantly improving the quality of the enhanced images and enhancing the required prominent lesion areas. These two methods are feasible for processing B-ultrasound images

, effective.

link: https://www.bxlimage.com/nw/367.html