Intellectual image intelligent recognition and sorting device and recognition and sorting method
A technology of intelligent identification and sorting device, applied in the field of intelligent image intelligent identification and sorting, intelligent image intelligent identification and sorting device, which can solve the problems of misjudgment and less than 100% identification, and achieve simple operation and calculation. , The effect of fast and accurate identification and sorting, saving labor costs
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0168] The method of the present invention is used for the identification that dynamic detection double gold has literal, and detection item and parameter are respectively: in original figure, utilize pixel size to be 752 * 480Pix; Literally through the dynamic detection and program, the edge of the boundary tracing and low grayscale connection image is obtained. In the boundary tracing and low grayscale connection mixture image, the standard deviation of the Gaussian filter is set to 3, and the maximum width of the detection edge is set to 11Pix; in the binarization diagram, the edge width is selected to be ≥ 3, and in the histogram, the detection method of the present invention is used to obtain the detection parameters of the double gold: the effective number of pixels is 56672Pix, the highest effective grayscale is 255Pix, and the lowest effective The grayscale is 94Pix; the high grayscale area SH is 42852Pix, the low grayscale area SL is 13820Pix, and the detection ratio K...
Embodiment 2
[0170] The method of the present invention is used for dynamic detection double gold no literal recognition detection, detection item and parameter are respectively: in original picture, utilize pixel size to be 752 * 480Pix; The frame rate is the CCD of MAX60F / S, to the double gold of continuous motion No Literal Through the dynamic detection and program, the edge of the boundary tracing and low grayscale connection image is obtained. In the boundary tracing and low grayscale connection mixture image, the standard deviation of the Gaussian filter is set to 3, and the maximum width of the detection edge is 7Pix ; In the binarized figure, the edge width selection ≥ 3, in the histogram, utilize the detection method of the present invention to obtain the detection parameters of double gold without literal: the effective pixel number is 56672Pix, the highest effective gray scale is 248Pix, the lowest effective gray The degree is 140Pix; the high gray area SH is 56464Pix, the low gr...
Embodiment 3
[0172] The method of the present invention is used for dynamic detection big white square silver dot reverse side recognition detection, and detection item and parameter are respectively: in original picture, utilize pixel size to be 752 * 480Pix; CCD of frame rate MAX60F / S, to continuous motion The back side of the big white square silver dot is dynamically detected and programmed to obtain the edge of the boundary tracing and low-gray-level mixed connection image. In the boundary-tracking and low-gray-level mixed connection image, the standard deviation of the Gaussian filter is set to 3, and the edge is detected. The maximum width is 8Pix; in the binary diagram, the edge width is selected to be ≥3, and in the histogram, the detection method of the present invention is used to obtain the detection parameters on the back of the silver point of the large white square: the number of effective pixels is 48825Pix, and the highest effective grayscale It is 240Pix, the minimum effec...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


