# US 5,710,837 A · Branch deletion method for thinned image > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 5,710,837 A - Document ID: US5710837A - Status: Granted - Kind: A - Application number: 08/459,988 - Filed: 1995-06-02 - Granted: 1998-01-20 - Earliest claimed: 1991-10-01 - Assignees: Yozan Inc. - Inventors: Koji Matsumoto; Manabu Oosaka - Web page: https://patentgrep.com/patent/US5710837A ## CPC - G06T 5/30: Erosion or dilatation, e.g. thinning - G06T 5/77: Retouching; Inpainting; Scratch removal - G06V 10/34: Smoothing or thinning of the pattern; Morphological operations; Skeletonisation - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract The present invention has an object to provide an image processing method for generating an image through thinning from which the topological characteristics of the image can be extracted more precisely and more accurately by deleting vestigial lines, that is, breaks in a line. ## Claims ### Claim 1 A method of removing a branch from a thinned image having a plurality of pixels, said method comprising the steps of: obtaining an electronic representation of said thinned image; electronically identifying a branch point and an end point of said electronic representation of said thinned image, said branch point being part of a main body of said electronic representation of said thinned image, said branch point connecting a branch with said main body; electronically separating said branch connected to said branch point from said main body, said branch being separated by deleting said branch point on said main body; electronically determining a length of said separated branch based on a number of pixels between said branch point and said end point in said electronic representation of said thinned image; electronically extracting said separated branch when said separated branch has a length equal to or less than a predetermined length; and electronically removing said extracted branch from said electronic representation of said thinned image. ### Claim 2 (depends on 1) The method of claim 1, said removing step comprising the steps of: electronically generating a mask image of only said branch; and electronically deleting said branch from said electronic representation of said electronic representation of said thinned image by masking said thinned image with said mask. ### Claim 3 (depends on 1) The method of claim 1, said identifying step comprising the step of locating said end point using a convolution operation. ### Claim 4 (depends on 1) The method of claim 1, further comprising the step of labelling a plurality of electronic representations corresponding to a plurality of thinned images in an image. ### Claim 5 (depends on 1) The method of claim 1, said separating step comprising a step of electronically deleting said branch point from said electronic representation of said thinned image. ### Claim 6 (depends on 1) The method of claim 1, further comprising the steps of: electronically deleting isolated points and minute lines from said electronic representation of said thinned image after electronically separating said branch from said main body; and electronically relabelling said thinned image after electronically deleting said isolated points and minute lines. ### Claim 7 (depends on 1) The method of claim 1, said extracting step comprising a step of electronically determining a length of said branch by generating a histogram of said line. ### Claim 8 (depends on 2) The method of claim 2, wherein: said generating step comprises the step of electronically inverting a portion of said thinned image containing said branch; and said deleting step comprises the step of electronically performing a bit-wise logical AND operation using said electronic representation of said thinned image and said electronically inverted image portion. ### Claim 9 (depends on 2) The method of claim 2, said deleting step comprises a step of using a multiplexer to selectively output portions of said thinned image based on said mask image. ### Claim 10 (depends on 1) The method of claim 1, wherein said removal step is performed using a mask based on the extracted branch. ### Claim 11 A method of removing a branch from a thinned image, said method comprising the steps of: obtaining an electronic representation of said thinned image; electronically identifying a main body and at least one branch in said electronic representation of said thinned image; generating a mask based on said at least one identified branch by: electronically separating said branch connected to a branch point from a main body of said electronic representation of said thinned image, said branch being separated by deleting said branch point on said main body; electronically determining a length of said branch; and electronically extracting said branch from said electronic representation of said thinned image when said branch has a length equal to or less than a predetermined length; and electronically removing said extracted branch from said electronic representation using said mask. ### Claim 12 (depends on 11) The method of claim 11, wherein said identification step includes the step of identifying a branch point and an end point of said electronic representation of said thinned image, said branch point being part of said main body, and said branch point connecting said branch with said main body. ### Claim 13 A method of removing a branch from a thinned image, said method comprising the steps of: obtaining an electronic representation of said thinned image; electronically identifying a branch point and an end point of said electronic representation of said thinned image, said branch point being part of a main body of said electronic representation of said thinned image, said branch point connecting a branch with said main body; electronically separating said branch connected to said branch point from said main body, said branch being separated by deleting said branch point on said main body; electronically determining a length of said separated branch based on a number of pixels between said branch point and said end point in said electronic representation of said thinned image; electronically generating a mask based on said separated branch when said separated branch has a length equal to or less than a predetermined length; and electronically removing a portion of said electronic representation of said thinned image corresponding to said mask.