Automatic system and method for inspecting hole quality

• US 4,555,798 A
• Filed: 06/20/1983
• Issued: 11/26/1985
• Est. Priority Date: 06/20/1983
• Status: Expired due to Fees

• Current Assignee: KLA Instruments, Corporation
• Original Assignee: KLA Instruments, Corporation
• Inventors: Wihl, Mark J., Buchholz, Steve, Eldredge, Peter, Broadbent, William H. Jr.
• Primary Examiner(s): Boudreau, Leo H.
• Application Number:
  US06/505,848
• Time in Patent Office: 890 Days
• Field of Search: 382/8, 382/22, 382/27, 382/28, 358/101, 358/106, 358/107, 356/237, 356/376, 356/379, 356/380, 356/384, 356/387, 356/394, 364/560, 364/563-564
• US Class Current: 382/144
• CPC Class Codes: G01N 21/95607 using a comparative method
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/30148 Semiconductor; IC; Wafer
• CPC Class Codes: G06T 7/0008 checking presence/absence
• CPC Class Codes: G06T 7/64 of convexity or concavity

Abstract

An automatic inspection system for inspecting holes in a mask including carriage means 30, illumination means 44, optical means 48, photosensitive detector means 46, and signal processing means 56. The mask 34 to be inspected is positioned by the carriage means in a horizontal plane. The optical means projects a focused image of a portion of the mask onto the photosensitive detector means. Photodiodes in the detector means are responsive to light from the illumination means that is transmitted through the holes in the mask. The signal processing means scans the outputs of the photodiodes and stores in memory a digital representation of the mask. The signal processing means performs inspection measurements and comparison tests. A smoothness checker circuit 240 measures the local radius of curvature of each hole at several places and compares the measurements to predetermined curvature limits to detect nicks and sharp protrusion defects. An area check circuit 246 measures the area of the hole and compares it to predetermined area limits. A diameter check circuit 244 measures the diameter of the hole in two dimensions and compares it to predetermined diameter limits. If either the hole area or the hole diameter is outside their respective limits, a hole size defect is indicated. Means are also provided for locating completely blocked, mispositioned or unintended holes by comparing detected information to a data base.

Claims

1. A method for inspecting the perimeter smoothness of a nominally round opening in an object comprising the steps of:

forming a two dimensional representation of the opening, said representation including two dimensional array of uniformly spaced pixels of a size substantially smaller than the size of said opening with each pixel having a range of digital values which indicate the presence, partial presence, or absence of said opening at the spatial position represented by that pixel, said array of pixels representing said opening as a connected group of pixels having first values at one end of said range surrounded by pixels having second values at the other end of the range, with the boundary therebetween representing the perimeter of said opening; and

examining the values of the pixels in one or more clusters surrounding said opening, each said cluster being composed of a group of several adjacent pixels arranged in a pattern with its center at said boundary, and determining that said perimeter is smooth at the location of a particular cluster if the number of pixels within said cluster having said first value is close to the number of pixels within said cluster having said second value, and wherein said perimeter is not smooth at said location if the number of pixels within said cluster having said first value is greatly disproportionate to the number of pixels within said cluster having said second value.

2. A method as recited in claim 1 for additionally inspecting the area of the opening in said object and further comprising the steps of:

counting the number of pixels in said array having said first value, said number being equal to an area count; and

setting an area error flag if said area count is less than a predetermined minimum value representing the smallest permissible area, or if said area count is greater than a predetermined maximum value representing the largest permissible area.

3. A method as recited in claim 1 for additionally inspecting the diameter of the opening in said object and further comprising the steps of:

counting the number of pixels in said array having said first value and spanning the distance from a first edge of said representation of said opening to a second edge thereof along one dimension to find a first measured diameter;

counting the number of pixels in said array having said first value and spanning the distance from a third edge of said representation of said opening to a fourth edge thereof along another dimension to find a second measured diameter; and

setting a diameter error flag if either said first measured diameter or said second measured diameter is less than a predetermined minimum value representing the smallest permissible diameter, or if either said measured diameter is greater than a predetermined maximum value representing the largest permissible diameter.

4. A method as recited in claim 1 wherein the axes of said two dimensional array are mutually perpendicular, said pixels are rectangular in shape, and said cluster is symmetrical about both axes.

5. A method as recited in claim 4 wherein said step of forming a two-dimensional representation of said object includes the steps of:

projecting an image of a portion of said object onto a one-dimensional column of uniformly spaced sensors, each said sensor being operable to detect the presence or absence of said image and to signal the presence of said image by generating a signal having a digital value of 1 and the absence of said image by generating a signal having a digital value of 0;

recording said signals in one column of a two-dimensional memory array;

moving said object in a direction perpendicular to the column length of said sensors to project an adjacent portion of said object onto said sensors;

recording the signals corresponding to said adjacent portion of said object in an adjacent column of said memory array;

repeating the previous two steps of moving said object and recording the signals for each adjacent portion of said object until at least a representation of a portion of said opening is contained in said memory array.

6. A method as recited in claim 5 wherein said object is moved a predetermined distance during each said moving step.

7. A method as recited in claim 5 wherein said object is moved at a constant velocity during said recording step.

8. A method as recited in claim 5 wherein said step of projecting an image of a portion of said object is accomplished by an optical lens mechanism, and wherein said sensors are photodiodes responsive to the intensity of light incident thereupon.

9. A method as recited in claim 8 wherein said object is opaque and is illuminated from the side opposite said sensors such that the illumination passes through said opening to said sensors.

10. A method as recited in claim 8 wherein said opening is a round aperture in an otherwise solid article and said object is illuminated from the side opposite said sensors.

11. A method as recited in claim 4 wherein said cluster of pixels includes a group of four pixels at the center thereof in a two-by-two square pattern, and wherein said cluster is determined to be positioned at said perimeter when two of the said four center pixels have values of 1 and the other two said center pixels have values of 0.

12. A method as recited in claim 11 wherein said cluster of pixels is an eight-by-eight symmetrical array with three pixels truncated at each corner thereof to form adjacent rows of four, six, eight, eight, eight, eight, six, and four pixels, respectively.

13. A method as recited in claim 3 for additionally detecting the complete blockage of said opening or the presence of an unintended opening, further comprising the steps of:

determining the coordinates of the center of said opening relative to a datum point;

comparing said coordinates to a predetermined set of reference coordinates; and

determining that said opening is blocked if no opening is encountered having coordinates within said predetermined tolerance of said reference coordinates; and

determining that said opening is misintended if said coordinates do not fall within a predetermined tolerance of said reference coordinates.

14. A method of inspecting one or more round openings in an object wherein each said opening is of the same nominal size, said method comprising the steps of:

forming a two dimensional representation of at least a portion of the object, said representation including a two-dimensional array of rows and columns of uniformly spaced apart pixels of a size substantially smaller than said openings with each pixel having a first digital value to indicate the presence of an opening at the spatial position of that pixel, or a second digital value to indicate the absence of an opening thereat, said array of pixels representing each opening as a connected group of pixels having said first value surrounded by pixels having said second value, with the boundary therebetween representing the perimeter of said opening;

storing said representation in a pixel memory;

scanning said pixel memory by accessing clusters of pixels therein one cluster at a time and summing the values of the pixels in each cluster as that cluster is accessed, said scanning step beginning at one corner of said array of pixels and proceeding to advance in the direction of the rows of pixels, each said cluster being composed of portions of several adjacent rows and columns of pixels arranged in a symmetrical pattern;

setting a smoothness defect flag in a chord memory array whenever an accessed cluster is positioned such that the center of said cluster is located at a boundary of one of said openings and the number of pixels within said cluster having said first value is a predetermined relationship to the number of pixels within said cluster having said second value; and

reporting a smoothness defect for an opening if any smoothness defect flags are set for that opening.

15. A method as recited in claim 14 additionally comprising the steps of:

setting a start chord flag in said chord memory array at a column corresponding to the position of an accessed cluster whenever the center of said cluster is positioned at said boundary and said start chord flag has not been previously set for said column;

incrementing a chord length register in said chord memory array at a column corresponding to the position of an accessed cluster whenever the center of said cluster is positioned inside the boundary of one of said objects;

setting an end chord flag in said chord memory array at a column corresponding to the position of an accessed cluster whenever the center of said cluster is positioned at said boundary and said start chord flag has been previously set for said column, said start chord flag and said end chord flag respectively designating the first and last pixels in a row of pixels that lies entirely within the boundaries of said representation of said object;

reporting an area defect for an opening if the sum of all of the chord length registers corresponding to that opening is less than a predetermined minimum area value or is greater than a predetermined maxium area value; and

reporting a diameter defect for an opening of either the largest value in said chord length register corresponding to that opening or the number of non-second value columns in said chord length register corresponding to that opening is less than a predetermined minimum diameter value or is greater than a predetermined maximum diameter value.

16. A method as recited in claim 15 wherein the axes of said two dimensional array are mutually perpendicular, said pixels are rectangular in shape, and said cluster is symmetrical about both axes.

17. A method as recited in claim 15 wherein said step of forming a two-dimensional representation of said openings includes the steps of:

projecting an image of a portion of said object onto a one-dimensional array of uniformly spaced sensors, each said sensor being operable to detect the presence or absence of said image and to generate a signal indicating the presence of said image at the sensor by a digital value of 1 and the absence of said image at the sensor by a digital value of 0;

recording the first group of signals generated by said sensors into one column of said pixel memory, each sensor defining a separate row in said pixel memory;

moving said object in a direction perpendicular to said sensors to project an adjacent portion of said objects onto said arrays of sensors;

recording the second group of sensor signals corresponding to said adjacent portion of said object into an adjacent column of said pixel memory;

repeating the previous two steps of moving said object and recording the subsequent groups of sensor signals for each adjacent portion of said object until a complete representation of said openings is contained in said pixel memory.

18. A method as recited in claim 17 wherein said forming and said scanning steps are performed concurrently by accessing one of said clusters after each said group of signals is recorded, said cluster including a present sub-column of pixels corresponding to the present position of said sensor, and trailing pixels previously recorded in the same rows as said present pixels.

19. A method as recited in claim 17 wherein said object is moved a discrete distance during said moving step and is stationary during said recording step.

20. A method as recited in claim 17 wherein said object is moved at a constant velocity during said recording step.

21. A method as recited in claim 17 wherein said step of projecting an image of a portion of said object is accomplished by an optical lens mechanism, and wherein said sensors are photodiodes responsive to the intensity of light incident thereupon.

22. A method as recited in claim 21 wherein said object is opaque and is illuminated from the side opposite said sensors such that the illumination passes through said openings to said sensors.

23. A method as recited in claim 21 wherein said openings are perforations in an otherwise solid article and said object is illuminated from the side opposite said sensors.

24. A method as recited in claim 16 wherein said cluster of pixels has a group of four pixels at the center thereof in a square pattern of two-by-two, and wherein said cluster is positioned at said boundary when two of the four center pixels have said first value and the other two center pixels have said second value.

25. A method as recited in claim 15 for additionally detecting the misposition of an opening and further comprising the steps of:

determining the coordinates of the center of said opening relative to a datum point;

comparing said coordinates to a predetermined set of reference coordinates; and

determining that said opening is mispositioned if said coordinates do not fall within a predetermined tolerance of said reference coordinates.

26. A method as recited in claim 16 for additionally detecting the blockage of an opening by determining that no set of coordinates falls within a predetermined tolerance of a set of reference coordinates corresponding to the intended location of an opening.

27. Inspection apparatus for inspecting the area, diameter, and surface smoothness of nominally round openings in an object comprising:

carriage means for supporting the object to be inspected and for simultaneously moving such object along an inspection path;

illumination means for illuminating portions of said object as it is moved along said path;

optical means for focusing an image of said object at a focal plane;

photosensitive detector means disposed at said focal plane at right angles to said inspection path for generating sensor signals corresponding to said image of said object incident thereon;

signal processing means for processing said sensor signals to inspect said openings, said signal processing means including memory means for storing said sensor signals, smoothness checking means for constructing a digital representation of portions of said object from said sensor signals and for measuring the radius of curvature of said openings, and for comparing said measured radius of curvature to an inspection standard, diameter checking means for measuring the diameters of said openings from said digital representation and for comparing said measured diameters to an inspection standard, area checking means for measuring the area of said openings from said digital representation and for comparing said measured areas to an inspection standard.

28. Inspection apparatus as recited in claim 27 and further including opening presence checking means for determining that an opening is not present if no opening is found to lie within a predetermined tolerance of an intended opening location.