US Patent No. 10,930,037

IMAGE PROCESSING DEVICE FOR DISPLAYING OBJECT DETECTED FROM INPUT PICTURE IMAGE


Patent No. 10,930,037
Issue Date February 23, 2021
Title Image Processing Device For Displaying Object Detected From Input Picture Image
Inventorship Yuta Namiki, Yamanashi (JP)
Fumikazu Warashina, Yamanashi (JP)
Assignee Fanuc Corporation, Yamanashi (JP)

Claim of US Patent No. 10,930,037

1. An image processing device comprising:a processor for selecting a model pattern of an object, and using the selected model pattern to detect one or more images of a plurality of objects from an input picture image in which the plurality of objects have been captured by a camera; and
a display controlled by the processor for graphically superimposing and displaying a detection result on each image thus detected,
wherein the display includes:
a first frame for displaying the entire input picture image in which the plurality of objects have been captured by the camera; and
a second frame for listing and displaying a plurality of partial picture images each including a single respective one of the plurality of objects separately cropped from the input picture image, such that each of the plurality of objects are displayed separately in the second frame, and wherein:
in the input picture image displayed in the first frame, a detection result determined based on a degree of match between the detected images and the model pattern is superimposed and displayed on all the detected images, in each of the plurality of partial picture images listed and displayed in the second frame, a respective detection result determined based on a degree of match between the each of the plurality of partial picture images and the model pattern is superimposed and displayed on an image corresponding to each of the plurality of partial picture images, and the first frame and the second frame are displayed simultaneously, wherein the display further superimposes and displays, on the partial picture image displayed in the second frame, first label information stored in association with an image corresponding to each partial picture image, the first label information indicates:
a match decision made by the processor when the degree of match between the detected images and the model pattern is larger than or equal to a threshold value, and
a non-match decision made by the processor when the degree of match between the detected images and the model pattern is less than the threshold, and
wherein the display allows an operator to:
perform a confirmation of the match decision and the non-match decision made by the processor by maintaining the first label information, and perform an override of the match decision and the non-match decision made by the processor by manually changing the first label information, and wherein when the operator performs the override, then the processor is configured to perform a learning process to update the model pattern based on the override of the match decision, the override of the non-match decision, the confirmation of the match decision, and the confirmation of the non-match decision as learning data.