Calibrate the starting line of fat thickness on the image and the method of fat thickness measurement
A fat thickness and image technology, applied in the computer field, can solve problems such as unusable methods, unguaranteed calculation accuracy, and unintuitiveness, and achieve the effect of facilitating implementation, high accuracy, and simple processing steps
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
example 1
[0072] The image of the present invention is an image obtained by ultrasonic equipment at the place where the body has fat. It is necessary to collect and process the ultrasonic image information first. To collect the ultrasonic image information, you can refer to the ultrasonic instrument device of the following Chinese patent CN209751086U, which can realize the acquisition of ultrasonic image information. And process ultrasound image information. refer to figure 1 .
[0073] This embodiment includes an ultrasonic transducer for sending out scanning beams, a digital control processing chip for controlling the ultrasonic transducer to send out scanning beams and collecting echo signals, and a digital control processing chip for sending control instructions to the digital control processing chip and viewing scanned images The portable control terminal, as well as the transmitting and receiving multiplexing circuit, the transmitting and receiving switching circuit, the transmit...
example 2
[0083] refer to figure 2 , the steps to calibrate the starting line of fat thickness on the image are as follows:
[0084] The first step is to import an image, and then start working on that image.
[0085] Perform Gaussian filtering on the imported image, and use a 3x3 Gaussian convolution kernel to perform convolution filtering on the image to blur the image, which is convenient for later gradient estimation and image segmentation processing. For the results, refer to image 3 .
[0086] 1. After Gaussian filtering, start to make image templates with strong edges, as follows:
[0087] (111). The Laplacian operator is used to calculate the edge image of the image, that is, the gradient image.
[0088] (112). The calculated edge image is absolute valued and normalized to be between 0-1. Specifically:
[0089] (113). A threshold value is used to perform threshold value processing on the edge image, and the threshold value can be set to 0.85. And after binarization, a bi...
example 3
[0101] refer to Figure 6 , when implementing Example 2, when performing the step of finding a connected domain whose length is greater than 50% of the image length, the preferred method is: enveloping the connected domain to obtain a circumscribing rectangle; calculating the length of the circumscribing rectangle. Firstly, the number of pixel columns occupied by the bounding rectangle is calculated, which is used for the total number of pixel columns of the entire image. For example, if the image has a total of 1080 pixel columns, and the circumscribed rectangular frame occupies 500 pixel columns, then this connected domain is not considered as a recognition area; while the image has a total of 1080 pixel columns, and the circumscribed rectangular frame occupies 550 pixel columns, then this connected domain is considered is the recognition area.
[0102]When implementing Example 2, when performing the step of finding a connected domain whose length is greater than 50% of the...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com