# US 10,311,577 B1 · Techniques to improve edge detection for images > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,311,577 B1 - Document ID: US10311577B1 - Status: Granted - Kind: B1 - Application number: 16/193,502 - Filed: 2018-11-16 - Granted: 2019-06-04 - Earliest claimed: 2018-11-16 - Assignees: Capital One Services, LLC - Applicants: Capital One Services, LLC - Inventors: Jeremy Edward Goodsitt; Austin Grant Walters; Fardin Abdi Taghi Abad; Kenneth Taylor; Reza Farivar; Vincent Pham; Anh Truong - Web page: https://patentgrep.com/patent/US10311577B1 ## CPC - G06T 7/13: Edge detection - G06T 7/90: Determination of colour characteristics - G06V 10/44: Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components - G06V 10/50: by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis - G06V 10/56: relating to colour - G06V 20/63: {Scene text, e.g. street names} - H04N 1/4092: {Edge or detail enhancement} - H04N 1/58: Edge or detail enhancement; Noise or error suppression, e.g. colour misregistration correction (H04N1/62 takes precedence) - H04N 1/60: Colour correction or control {(H04N1/54 takes precedence)} - H04N 1/6002: {Corrections within particular colour systems} - G06T 2207/10024: Color image - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract Techniques to improve edge detection in images. Some techniques include logic to process image data into patched image data in accordance with a colorspace model where the patched image data includes color data in a plurality of patches and identify an image group corresponding to the patched image data. The logic may be further configured to select, based upon the image group, a colorspace transform mechanism being operative to transform the image data into transformed image data in accordance with another colorspace model, the other colorspace model having a higher likelihood than the colorspace model at edge detection for the image group. The logic may be further configured to apply the colorspace transform mechanism to the image data to generate the transformed image data in accordance with the other colorspace mode and then, apply an edge detection technique to the transformed image data. Other embodiments are described and claimed. ## Claims ### Claim 1 An apparatus, comprising: a processing circuit; and logic stored in computer memory and executed on the processing circuit, the logic to cause the processing circuit to: process image data into patched image data, the image data being configured in accordance with a colorspace model, the patched image data comprising a plurality of patches of which each patch comprises color data; identify an image group corresponding to the patched image data; select a colorspace transform mechanism based upon the image group, the colorspace transform mechanism being operative to transform the image data into transformed image data in accordance with another colorspace model, the other colorspace model having a higher likelihood than the colorspace model at edge detection for the image group; apply the colorspace transform mechanism to the image data to generate the transformed image data in accordance with the other colorspace model; and apply an edge detection technique to the transformed image data. ### Claim 2 (depends on 1) The apparatus of claim 1 further comprising logic operative to cause the processing circuit to transform the image data comprising a red channel, a green channel, and a blue channel into the transformed image data comprising a luminance channel, an alpha channel, and a beta channel. ### Claim 3 (depends on 1) The apparatus of claim 1 further comprising logic operative to cause the processing circuit to process a collection of images and converting each image into a grid of patches of which each patch comprises histogram data. ### Claim 4 (depends on 3) The apparatus of claim 3 further comprising logic operative to cause the processing circuit to cluster the collection into groups of images of which each image group corresponds to a particular colorspace transform for substantially maximizing edge detection. ### Claim 5 (depends on 3) The apparatus of claim 3 further comprising logic operative to cause the processing circuit to: for each image group, select a sample set of images, compare mean histogram data of the sample set of images with a mean histogram data of the image to produce a comparison result, and cluster the collection into groups of images based upon the comparison result. ### Claim 6 (depends on 1) The apparatus of claim 1 further comprising logic operative to cause the processing circuit to recognize text in the transformed image. ### Claim 7 (depends on 1) The apparatus of claim 1 further comprising logic operative to cause the processing circuit to apply a mini-colorspace transform for at least one patch of the patched image data. ### Claim 8 A computer-implemented method executed on at least one processing circuit, comprising: process image data into patched image data, the image data being configured in accordance with a colorspace model, the patched image data comprising a plurality of patches of which each patch comprises color data; identify an image group corresponding to the patched image data; select a colorspace transform mechanism based upon the image group, the colorspace transform mechanism being operative to transform the image data into transformed image data in accordance with another colorspace model, the other colorspace model having a higher likelihood than the colorspace model at edge detection for the image group; apply the colorspace transform mechanism to the image data to generate the transformed image data in accordance with the other colorspace model; and apply an edge detection technique to the transformed image data to determine whether the image data includes sensitive information. ### Claim 9 (depends on 8) The computer-implemented method of claim 8, comprising transforming the image data comprising a red channel, a green channel, and a blue channel into the transformed image data comprising a luminance channel, an alpha channel, and a beta channel. ### Claim 10 (depends on 8) The computer-implemented method of claim 8, comprising processing a collection of images and converting each image into a grid of patches of which each patch comprises histogram data. ### Claim 11 (depends on 10) The computer-implemented method of claim 10, comprising clustering the collection into groups of images of which each image group corresponds to a particular colorspace transform for substantially maximizing edge detection. ### Claim 12 (depends on 10) The computer-implemented method of claim 10, comprising: for each image group, selecting a sample set of images, comparing mean histogram data of the sample set of images with a mean histogram data of the image to produce a comparison result, and clustering the collection into groups of images based upon the comparison result. ### Claim 13 (depends on 8) The computer-implemented method of claim 8, comprising recognizing text in the transformed image data. ### Claim 14 (depends on 8) The computer-implemented method of claim 8, comprising applying a mini-colorspace transform for at least one patch of the patched image data. ### Claim 15 At least one non-transitory computer-readable storage medium comprising instructions that, when executed, cause a system to: process image data into patched image data, the image data being configured in accordance with a colorspace model, the patched image data comprising a plurality of patches of which each patch comprises color data; identify an image group corresponding to the patched image data; select a colorspace transform mechanism based upon the image group, the colorspace transform mechanism being operative to transform the image data into transformed image data in accordance with another colorspace model, the other colorspace model having a higher likelihood than the colorspace model at edge detection for the image group; apply the colorspace transform mechanism to the image data to generate the transformed image data in accordance with the other colorspace model; and apply an edge detection technique to the transformed image data to determine whether the image data includes sensitive information. ### Claim 16 (depends on 15) The non-transitory computer-readable storage medium of claim 15, comprising instructions that when executed cause the system to: process a collection of images and converting each image into a grid of patches of which each patch comprises histogram data. ### Claim 17 (depends on 15) The non-transitory computer-readable storage medium of claim 15, comprising instructions that when executed cause the system to: cluster the collection into groups of images of which each image group corresponds to a particular colorspace transform for substantially maximizing edge detection. ### Claim 18 (depends on 15) The non-transitory computer-readable storage medium of claim 15, comprising instructions that when executed cause the system to: for each image group, select a sample set of images, comparing mean histogram data of the sample set of images with a mean histogram data of the image to produce a comparison result, and clustering the collection into groups of images based upon the comparison result. ### Claim 19 (depends on 15) The non-transitory computer-readable storage medium of claim 15, comprising instructions that when executed cause the system to: recognize text in the transformed image. ### Claim 20 (depends on 15) The non-transitory computer-readable storage medium of claim 15, comprising instructions that when executed cause the system to: apply a mini-colorspace transform for at least one patch of the patched image data.