Method and device for positioning characters in image
A positioning method, a technology in images, applied in character and pattern recognition, instruments, computer components, etc., can solve problems such as inaccurate judgment results, unknown background complexity of text areas, and lack of universal applicability, and achieve positioning accurate results
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0042] figure 2A text positioning method in an image provided by this embodiment is shown. Wherein, in this embodiment, when performing DCT transformation on image blocks, firstly, each image block is divided into several sub-blocks, and then the sub-blocks are respectively subjected to DCT transformation. Correspondingly, when calculating the texture energy difference degree value of an image block, also Firstly, the texture energy of each sub-block in the image block is calculated, and then the texture energy difference degree value of the image block is calculated according to the texture energy of each sub-block. Combine below figure 2 This embodiment will be described in detail.
[0043] Such as figure 2 As shown, the text positioning method in the image provided by this embodiment includes:
[0044] Step S201: Divide the image into several image blocks, and then divide each image block into several sub-blocks, perform DCT transformation on each sub-block in each i...
Embodiment 2
[0075] In this embodiment, the specific implementation of the second embodiment is described in detail by taking an image block with a size of 16×16 pixels and a sub-block with a size of 8×8 pixels as an example.
[0076] Such as Figure 5 As shown, the text positioning method in the image provided by this embodiment includes:
[0077] Step S501: Divide the image into image blocks in units of 16x16 pixels, and then divide each 16x16 pixel image block into four sub-blocks of 8x8 pixels. The DCT transformation is performed on the luminance space, Cb space and Cr space of each sub-block of 8x8 pixels. The specific method of DCT transformation can refer to relevant materials, and will not be repeated here. Compared with the prior art, this embodiment comprehensively considers the Cb value and Cr value of luminance information and chromaticity space, which can reflect the information of the image more comprehensively, thereby providing more detailed information for distinguishing ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 