In the process of animal B-ultrasound image processing, a method of detecting the maximum value of wavelet transform mode was proposed in the measurement method of animal B-ultrasound image. Since then, wavelet transform has been widely used in image processing field. Some people have applied it to the edge extraction of echocardiography and the enhancement of animal B-ultrasound images. However, wavelet transform also has its inherent shortcomings, which are manifested as poor signal-to-noise ratio on small scale and inaccurate positioning on large scale, so there are uncertainties in edge detection on all scales. When using wavelet transform for edge detection, it is necessary to balance the contradiction between effective detection and accurate positioning.

Veterinary B-ultrasound BXL-DZ20

Fuzzy set theory has been applied in the field of veterinary B-ultrasound image processing in recent years, and its advantage is that it is good at dealing with uncertainty problems. It combines fuzzy set theory and wavelet transform multi-scale information to detect the edge, and forms a fuzzy multi-scale edge detection method. Some people apply FEMD to the left ventricular volume detection of two-dimensional veterinary B-ultrasound echocardiography, and put forward the maximum value theory, and then apply fuzzy set theory to the left ventricular edge detection, which is necessary to apply to the veterinary B-ultrasound.

The mode local maximum edge method considers that the mode local maximum point in the gradient direction of wavelet transform is the point with drastic change of image gray level, that is, the edge point. Therefore, the wavelet transform results of each scale can be regarded as a fuzzy set, which represents the degree to which the points in the animal B-ultrasound image belong to the edge point set. A maximum edge detection method is proposed for the case where there is only a single step edge in the signal to be detected.

link: https://www.bxlimage.com/nw/254.html