Method and device for conveying object recognition based on image separation
An object recognition and image technology, applied in the field of image recognition, can solve the problems of high cost and high error
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0029] Such as figure 1 Shown, based on the image separation method for handling object recognition, the method performs the following steps:
[0030] Step 1: Obtain the original image containing the object to be transported, and copy an original image to obtain the copied original image;
[0031] Step 2: Separating the foreground image from the original image to obtain a foreground image and a background image respectively; at the same time, separating the foreground image from the copied original image to obtain a copied foreground image and a copied background image respectively;
[0032] Step 3: performing image fusion on the foreground image and copying the foreground image to obtain the fused foreground image;
[0033] Step 4: Perform image difference between the background image and the copied background image to obtain the background image after difference;
[0034] Step 5: Obtain the position of the pixel point whose pixel value is not 0 in the background image afte...
Embodiment 2
[0039] On the basis of the previous embodiment, the step 2: separating the foreground image from the original image to obtain the foreground image and the background image respectively includes: determining that each pixel in the original image belongs to the foreground pixel of the foreground pixel first Prior probability, prior probability of background pixels belonging to background pixels; according to the prior probability of foreground pixels and the prior probability of background pixels of each pixel, determine the pixel type of each pixel, and the pixel type includes Foreground pixel points, background pixel points, and mixed pixel points; according to each of the mixed pixel points, the prior probability of the foreground pixel points of each of the described mixed pixel points, and the background image, the fusion weight of each of the mixed pixel points is determined value, and determine the foreground component value of each mixed pixel according to the fusion weig...
Embodiment 3
[0043] On the basis of the previous embodiment, according to the original image, the background image, and the prior probability model of pixels, determine the prior probability of each pixel in the original image belonging to the foreground pixel, and the prior probability of the foreground pixel belonging to the background. The background pixel prior probability of the pixel point; the determination method of the pixel point prior probability model includes: obtaining a sample image and the background image, and the sample image includes the original image in the video stream where the original image is located. The image of the previous video frame; according to the pixel value of the sample image and the pixel value of the background image, threshold filtering is performed on the sample image, and seed foreground pixels and seed background pixels are extracted from the sample image ; According to the extracted seed foreground pixel and seed background pixel, establish the p...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More