Method for extracting target area in inner scene video of oven on basis of level set

Abstract

The invention discloses a method for extracting a target area in an inner scene video of an oven on the basis of a level set. By means of a mask technique, an improved level set model and a region threshold value technique, a foreground area where target colors are consistent in an inner scene of a chamber is quickly extracted. The method includes the following steps that (1) a video of cooking in the chamber of the oven is shot, and a collection interval is set so that video frame images needing to be processed are obtained; (2) an interesting area is circled out from an image at the first frame, a sub-image segmentation mask is generated for obtaining sub pictures; (3) cutting is carried out andsub picturesare formed; (4) an initial outline curve is set; (5) level set evolution is carried out; (6) threshold processing is carried out; (7) a target is extracted; (8) the image at the next frame is read, and the third step to the seventh step are carried out again until target extraction of the images of all the frames is finished. According to the method, a clear target boundary can be obtained, one frame image only needs to be iterated for 6-12 times, and online detection efficiency is very high.

Description

technical field [0001] The invention belongs to the field of scene video processing in an oven, and in particular relates to a method for extracting a target region in a video frame image, which can be used for target extraction of grilled food in a universal steam oven. Background technique [0002] In the process of steaming and roasting food in a universal steaming oven, it is necessary to identify the food area in the scene in the oven, so as to automatically detect the color quality and ensure the quality of steaming and baking. However, the technology of automatically extracting the target area in the scene in the oven hard to accomplish. Although the level set evolution curve has the function of automatically extracting the target area of ​​free boundary shape[Li C et al.Distance regularized level set evolution and its application to image segmentation[J].Image Processing,IEEE Transactions on.2010,19(12): 3243-3254], however, when encountering the weakened target bou...

AI Technical Summary

Problems solved by technology

[0002] In the process of steaming and roasting food in a universal steaming oven, it is necessary to identify the food area in the scene in the oven, so as to automatically detect the color quality and ensure the quality of steaming and baking. However, the technology of automatically extracting the target area in the scene in the oven hard to accomplish

Although the level set evolution curve has the function of automatically extracting the target area of ​​free boundary shape[Li C et al.Distance regularized level set evolution and its application to image segmentation[J].Image Processing,IEEE Transactions on.2010,19(12): 3243-3254], however, when encountering the weakened target boundary in the video image of the scene in the oven, it is easy to cause the evolution to cross the boundary; and the number of iterations of the level set evolution of each frame image needs more than 200 times, if the online detection is implemented, the efficiency is very high low

Images