# US 4,958,373 A · Defect-recognition processing apparatus > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 4,958,373 A - Document ID: US4958373A - Status: Granted - Kind: A - Application number: 07/174,879 - Filed: 1988-03-29 - Granted: 1990-09-18 - Earliest claimed: 1987-03-31 - Assignees: Kabushiki Kaisha Toshiba - Inventors: Toshiro Usami; Hiroyuki Kamijo; Takao Ohta; Masanobu Ogino - Web page: https://patentgrep.com/patent/US4958373A ## CPC - G06T 7/0004: {Industrial image inspection} - G01N 21/9501: {Semiconductor wafers (manufacturing processes per se of semiconductor devices implementing a measuring step H10P74/20)} - G01N 21/8851: {Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges (G01N21/8806 and G01N21/93 - G01N21/95692 take precedence; optical measurement of dimensions G01B11/00; optical scanning G02B26/10; image transformation G06T3/00; computerised image enhancement G06T5/00; image processing per se for flaw detection G06T7/0002)} - G01N 2021/8874: {Taking dimensions of defect into account} - G01N 2021/8887: {based on image processing techniques} - G01N 2201/102: Video camera - G06T 2207/10056: Microscopic image - G06T 2207/30148: Semiconductor; IC; Wafer - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q) ## Abstract A defect-recognition processing apparatus converts into a defect image pattern, via a television camera, crystal defects present on the surface of an object under inspection, to process an image signal, by means of an image processing device, which corresponds to the defect image pattern, to measure rectangular images in terms of their length and their ratio between L.sub.Y and L.sub.X (L.sub.Y : a length in a longitudinal direction and L.sub.X : a length in the lateral direction of the wafer) and to detect defects developed on the surface of the aforementioned object. ## Claims ### Claim 1 A defect-recognition processing apparatus comprising: image converting means for converting, into an image, surface data corresponding to a predetermined area on a surface of an object being inspected; image processing means for processing the image from said image converting means and for detecting the presence of a defect image having both a length greater than a predetermined value and a ratio L.sub.y /L.sub.X having a value greater than n or a value lower than 1/n where L.sub.Y denotes a length of the defect image in a longitudinal direction and L.sub.X denotes a length of the defect image in a lateral direction; and system control means for controlling operations performed by said image processing means. ### Claim 2 (depends on 1) A defect-recognition processing apparatus according to claim 1, wherein said object is a semi-conductor wafer. ### Claim 3 (depends on 1) A defect-recognition processing apparatus according to claim 1, wherein said rectangular image has a length greater than 2 .mu.m and said ratio L.sub.Y /L.sub.X is in a range of a value greater than 1.6 or lower than 0.6. ### Claim 4 (depends on 1) A defect-recognition processing apparatus according to claim 1, wherein a horizontal TV scanning direction corresponds to a direction parallel to or perpendicular to one side of said rectangular image. ### Claim 5 (depends on 1) A defect-recognition processing apparatus according to claim 1, wherein said object is a silicon wafer having a (100) face as a major surface and an orientation flat in a <110> direction and only those defects longer in the <110> or the equivalent direction are detected. ### Claim 6 (depends on 5) A defect-recognition processing apparatus according to claim 5, wherein a horizontal TV scanning is performed in a direction horizontal to or perpendicular to said orientation flat. ### Claim 7 (depends on 5) A defect-recognition processing apparatus according to claim 5, wherein those defects longer in said <110> direction are detected to find a defect density, and a signal corresponding to double said defect density is output as a crystal defect density. ### Claim 8 (depends on 1) A defect-recognition processing apparatus according to claim 1, wherein said object has a (111) face as a major surface and only those defects longer in the <110>, or the equivalent direction are detected. ### Claim 9 (depends on 5) A defect-recognition processing apparatus according to claim 5, wherein those defects longer in said <110> direction are detected to obtain a defect density, and an output corresponding to three times said defect density is delivered as a crystal defect density. ### Claim 10 (depends on 1) A defect-recognition processing apparatus according to claim 1, wherein said object is a silicon wafer having a stepped pattern on the surface, and said image processing means converts the picked-up TV signal to a binary signal and detects defects after elimination of an image pattern formed by said stepped pattern. ### Claim 11 (depends on 1) A defect-recognition processing apparatus according to claim 1, wherein said image converting means illuminates a flat surface of said object with light coming diagonally from above, and receives reflected light to be picked up as TV data. ### Claim 12 (depends on 1) A defect-recognition processing apparatus according to claim 1, wherein said object is a silicon wafer having a (100) face as a major surface and light is received which is reflected from the (100) face of the silicon wafer surface. ### Claim 13 (depends on 12) A defect-recognition processing apparatus according to claim 12, wherein light is incident on said (100) face of said silicon wafer at an angle of 54.degree.<.theta.<90.degree., and is reflected, at an angle of 108.degree.-.theta., on that light reflection face. ### Claim 14 (depends on 1) A defect-recognition processing apparatus according to claim 1, wherein said image processing means detects an image signal, as said picked-up TV signal, in the form of a shade image and an image different in brightness level from a background as a stacking fault.