Text recognition method and apparatus

A text recognition apparatus, comprising:

• at least one processor;
• and
• one or more memories coupled to the at least one processor and storing programming instructions for execution by the at least one processor to cause the apparatus to:
• obtain a to-be-detected image;
• determine a target text detection area in the to-be-detected image, wherein the target text detection area comprises target text in the to-be-detected image, the target text detection area is a polygonal area, the polygonal area comprises m vertex pairs, m is a positive integer greater than 2, and each of the m vertex pairs comprises two vertices including a vertex located on one side of the target text and another vertex located on another side of the target text, so that m vertices are located on one side of the target text, and other m vertices are located on another side of the target text;
• correct the polygonal area to m−1 rectangular areas to obtain a corrected target text detection area based on the m vertex pairs;
• and
• perform text recognition on the corrected target text detection area to determine the target text, and output the target text,
• wherein the determining of the target text detection area in the to-be-detected image comprises:
• determining a plurality of candidate center points in the to-be-detected image and a vertex pair corresponding to each of the plurality of candidate center points, wherein each of the plurality of candidate center points is associated with a confidence level indicating likelihood of the associated candidate center point being a center point in the target text detection area in a text height direction;
• and
• determining, based on the confidence level of each of the plurality of candidate center points, center points forming a center line of the target text detection area, wherein the center line passes though all pieces of text in the target text detection area.

A non-transitory storage medium storing computer program code, wherein the computer program code comprises computer instructions, and when the computer instructions are executed by at least one processor, the at least one processor is enabled to perform operations comprising:

• obtaining a to-be-detected image;
• determining a target text detection area in the to-be-detected image, wherein the target text detection area comprises target text in the to-be-detected image, the target text detection area is a polygonal area, the polygonal area comprises m vertex pairs, m is a positive integer greater than 2, and each of the m vertex pairs comprises two vertices including a vertex located on one side of the target text and another vertex located on another side of the target text, so that m vertices are located on one side of the target text, and other m vertices are located on another side of the target text;
• correcting the polygonal area to m−1 rectangular areas to obtain a corrected target text detection area based on the m vertex pairs;
• and
• performing text recognition on the corrected target text detection area to determine the target text, and outputting the target text,
• wherein the determining of the target text detection area in the to-be-detected image comprises:
• determining a plurality of candidate center points in the to-be-detected image and a vertex pair corresponding to each of the plurality of candidate center points, wherein each of the plurality of candidate center points is associated with a confidence level indicating likelihood of the associated candidate center point being a center point in the target text detection area in a text height direction;
• and
• determining, based on the confidence level of each of the plurality of candidate center points, center points forming a center line of the target text detection area, wherein the center line passes though all pieces of text in the target text detection area.