Text detection method, device, electronic device and computer storage medium
A text detection and text technology, applied in the computer field, can solve the problems of cost, time-consuming, large computing resources, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0034] Embodiment 1 of the present application provides a text detection method, such as figure 1 as shown, figure 1 It is a flow chart of a text detection method provided in the embodiment of the present application, and the text detection method includes the following steps:
[0035] Step 101, perform feature extraction on the text image to be detected, and obtain a real text probability map and at least one pixel class probability map corresponding to the text image to be detected.
[0036] It should be noted that the text detection method in the embodiment of the present application is applicable to text detection with various text densities, including but not limited to regular density text, dense density text, sparse density text, especially dense density text. Among them, specific indicators for determining whether a certain text is a dense text can be appropriately set by those skilled in the art according to the actual situation, including but not limited to: accordi...
Embodiment 2
[0052] Optionally, in an embodiment of the present application, step 103 may further include step 103a and step 103b.
[0053] Step 103a, according to at least one pixel class probability map, determine the pixel class of each pixel in the text image to be detected.
[0054] Take a text image to be detected that includes pixels of four pixel categories as an example. The probability map of each pixel category indicates the probability that each pixel in the text image to be detected belongs to the category. For example, the pixel to be detected The text image includes 200 pixels, and the first type pixel class probability map indicates the probability that 200 pixel points belong to the first type pixel point, that is, the probability that these 200 pixel points are located in a non-overlapping area. Similarly, the second to fourth pixel category probability maps represent the probabilities of 200 pixels belonging to the second to fourth categories of pixels respectively. Tha...
Embodiment 3
[0061] Optionally, in an embodiment of the present application, step 105 may further include step 105a1-step 105a3.
[0062] Step 105a1: Calculate connected domains for at least one pixel-type binary image except for the reference pixel-type binary image to obtain at least one candidate connected domain.
[0063] Taking the binary image of four pixel categories as an example, a binary image of a pixel category contains at least one text area, and the remaining binary images of the second to fourth pixel categories are all connected domains , the second to fourth connected domains can be obtained, and the second to fourth connected domains are used as connected domains to be selected, that is, at least one connected domain includes the second connected domain, the third connected domain and the fourth connected domain Class connected domain. In step 105a1, when obtaining at least one connected domain to be selected, it can be processed in parallel, that is, to obtain the conne...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


