# US 2003/0210818 A1 · Knowledge-based hierarchical method for detecting regions of interest > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2003/0210818 A1 - Document ID: US20030210818A1 - Status: Published application - Kind: A1 - Application number: 10/142,175 - Filed: 2002-05-09 - Published: 2003-11-13 - Earliest claimed: 2002-05-09 - Assignees: General Dynamics C4 Systems, Inc. - Inventors: Glen P. Abousleman; Huibao Lin; Jennie Si - Web page: https://patentgrep.com/patent/US20030210818A1 ## CPC - G06V 10/25: Determination of region of interest [ROI] or a volume of interest [VOI] - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A knowledge-based hierarchical method for detecting regions of interests (ROIs) uses prior knowledge of the targets and the image resolution in detecting ROIs. The result produces ROIs that contain only one target that is completely enclosed within the ROI. The detected ROI can conform to the shape of the target even if the target is of irregular shape. Furthermore, the method works well with images that contain connected targets or targets broken into pieces. The method is not sensitive to contrast levels and is robust to noise. Thus, this method effectively detects ROIs in common real world imagery that has a low resolution without costly processing while providing fast and robust results. ## Claims ### Claim 1 A method for detecting a region of interest in an image, comprising the steps of: using prior knowledge to identify potential objects of interest in the image; generating a plurality of down-sampled images from the image; detecting the potential objects of interest in the plurality of down-sampled images; and combining all down-sampled images wherein each potential object of interest is identified by a region of interest. ### Claim 2 (depends on 1) The method of claim 1 further comprising the step of removing unnecessary objects of interest in each down-sampled image before said combining step. ### Claim 3 (depends on 1) The method of claim 1 wherein the prior knowledge step comprises the step of using prototype information and sensor information. ### Claim 4 (depends on 1) The method of claim 1 wherein the generating step comprises the step of using a wavelet transform. ### Claim 5 (depends on 1) The method of claim 1 wherein the detecting step comprises the step of identifying the potential object of interest by its edge. ### Claim 6 (depends on 1) The method of claim 1 wherein the using step comprises the steps of: grouping objects of interest based on prototype information provided by prior knowledge; and designing a detection algorithm for each group of objects to be used in said detecting step. ### Claim 7 (depends on 6) The method of claim 6 further comprising the step of splitting at least two objects of interest after said detecting step. ### Claim 8 (depends on 7) The method of claim 7 further comprising the step of merging an object of interest after said splitting step. ### Claim 9 (depends on 1) The method of claim 1 wherein said combining step comprises the step of using a multi-voting algorithm. ### Claim 10 (depends on 8) The method of claim 8 further comprising the step of performing probability assignments on each region of interest to modify the region of interest to conform to the shape of the object of interest. ### Claim 11 (depends on 1) The method of claim 1 further comprising the step of expanding the region of interest. ### Claim 12 A method for detecting a region of interest in an image containing objects, comprising the steps of: dividing prototypes into at least one group based on prior knowledge of the image; generating at least one copy of the image at the same resolution of the image or lower; detecting a plurality of regions that contains an object in each group in at least one copy of the image; consolidating the plurality of regions; and expanding the consolidated region to depict at least one region of interest. ### Claim 13 (depends on 12) The method of claim 12 wherein the expanding step comprises depicting one object in each region of interest. ### Claim 14 (depends on 12) The method of claim 12 further comprising the step of using a feature of the group of objects to aid in the detecting step. ### Claim 15 (depends on 12) The method of claim 12 wherein the detecting step, the consolidating step, and the expanding step each comprise the step of using prior knowledge. ### Claim 16 (depends on 12) The method of claim 12 wherein the detecting step further comprises the steps of: splitting a region in the plurality of regions; and merging at least two regions in the plurality of regions. ### Claim 17 (depends on 12) The method of claim 12 wherein the consolidating step comprises the step of using an arbitration process. ### Claim 18 (depends on 12) The method of claim 12 wherein the detecting step comprises the step of performing the detection of an object in a copy of the image in parallel with the detection of another object in a copy of the image. ### Claim 19 (depends on 12) The method of claim 12 further comprising the step of using prior knowledge to aid in said detecting step. ### Claim 20 The method of 13 further comprising the step of performing probability assignments on each region of interest to modify the region of interest to conform to the shape of the object. ### Claim 21 (depends on 12) The method of claim 12 further comprising the step of designing a detection algorithm for each group of objects to be used in said detecting step.