Picture identification and labeling method
A picture recognition and picture technology, applied in the field of picture processing, can solve the problems of inconvenient picture processing, unable to mark pictures, etc., and achieve the effect of reducing labor load
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0034] Such as Figure 1 to Figure 3 A picture recognition and labeling method shown, including picture recognition and picture labeling;
[0035] Image recognition includes the following steps:
[0036] a1. Obtain the image to be identified and the image pre-stored in the system;
[0037] a2. Preprocessing of the picture to be recognized: longitudinal scanning and horizontal scanning are respectively performed on the picture to be recognized to obtain a vertical pixel set and a horizontal pixel set;
[0038] a3. Compare the preset information between the vertical pixel set and the horizontal pixel set and the system pre-stored picture, and obtain the characteristic image of the picture to be recognized;
[0039] a4. Carry out partition processing on the feature image, and obtain pictures after multiple partitions;
[0040] a5. Perform image processing on the partitioned image to obtain the image number after comparison;
[0041] Image annotation includes the following ste...
Embodiment 2
[0049] When the image to be identified is an engineering drawing, it is necessary to identify the specific dimensioning content;
[0050] According to national standards, there are many ways to express dimensions according to different types, such as the dimensions that express length and width separately. At this time, they are only composed of numbers. See Figure 4 A of A; or when expressing the diameter, it is composed of symbols and numbers, see Figure 4 B; or also include dimension tolerances, see Figure 4 C of ; Since these dimensions are standard, the system pre-stored pictures related to dimensioning can be formed through the above-mentioned national standards.
[0051] These pre-stored images in the system are divided into three partitions, namely Partition 1, Partition 2 and Partition 3. Partition 1 is mainly for various symbols, Partition 2 is for numbers, and Partition 3 can be single-line or single-line according to the tolerance size requirements. is a doubl...
Embodiment 3
[0056] This embodiment is a preferred solution of Embodiment 2, that is, after the identification is completed, the specific identification content is marked; the significance of the marking is that these identified size data need to be summarized, and each size is labeled in numerical order Finally, the data exported to excel will also have a one-to-one corresponding sequence number, so that the specific data can be found conveniently and quickly.
[0057] see Figure 5 , after completing the marking, a box will be automatically added to the dimensioning content, and the specific serial number will be marked near the box; when marking, proceed as follows:
[0058]Take the specific dimensioning content as the center to form a dimensioning frame; assuming that the dimensioning content is a rectangle, then the border of the formed dimensioning frame must be larger than the largest side of the rectangle to ensure that it can be marked effectively;
[0059] Then the area between ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More