DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments/Amendments
Applicant's arguments filed on March 30, 2026 have been fully considered but they are not persuasive.
Applicant argues in page 8 of the remarks, that “the centroid of the "every object" is merely a location of an object that matches with a template image, not corresponding to the unit patterns recited in claims 1 and 13 of the instant case”, and that “the present application can obtain the center position of each pattern, without using edge information of pattern”. The examiner respectfully disagrees.
Chen (Chen, H.-C. (2020). Automated Detection and Classification of Defective and Abnormal Dies in Wafer Images. Applied Sciences, 10(10), 3423. https://doi.org/10.3390/app10103423) is only relied upon for the steps of blurring each of the unit patterns, which are the die patterns, and calculating respective center positions based on the blurring. Additionally, claims 1, and 13 do not recite the feature of obtaining center positions of each pattern, without using edge information of pattern. Bhaskar already teaches extracting and aligning the unit patterns (structures images 46) for averaging, and combining Bhaskar and Chen would have been obvious to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention and apply blurring, and finding the centroids of each unit pattern based on the blurring, improving Bhaskar’s extraction of the unit patterns (structures corresponding to images 46).
The reason for doing so is “to predict the locations of pattern candidates that possibly contain certain predefined patterns from the detected die patterns and their spatial properties” (Chen page 3, 2nd bullet.).
Applicant further argues in page 8 of the remarks, that “the centroid of glyph image of Nolan is merely a center of mass of the glyph image”. The examiner respectfully disagrees.
Under the broadest reasonable interpretation, Nolan’s glyph images are unit patterns. Nolan (US 20070189628 A1) is only relied upon calculating centroids and setting the centroids of images of unit patterns (glyph images) as reference positions so they can be stacked with precise alignment based on the centroids as described in Nolan paragraphs 47-48. It would have been obvious to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention to combine Bhaskar and Nolan, to find centroids of unit patterns.
The reason for doing so is because the “images are precisely aligned and overlaid so that the entire cluster is arranged in one large stack” (Nolan: ¶48).
Regarding the amended limitation with respect to claims 1,and 13: “splitting the full image into a plurality of unit images based on the respective reference positions for each unit image of the plurality of unit images to include one unit pattern of the unit patterns”;
Bhaskar et al. (US 20090041332 A1) teaches:
splitting the full image into a plurality of unit images based on the respective reference positions for each unit image of the plurality of unit images to include one unit pattern of the unit patterns (Bhaskar: Fig. 8-9, ¶130-131, “The multiple images are acquired at multiple positions on the wafer at which the structure is formed. In one embodiment, the multiple positions include positions of the structure in cells having identical designs. . . multiple dies from each focus and exposure point can be averaged together. . . individual structures may have random variations. . . by averaging multiple images of the structure together, either from identical cells (e.g., adjacent identical cells) or from neighboring dies, a composite image can be constructed with much lower LER or random variation than any individual structure. . .”; NOTE: The full image is the output of the inspection system which is an image of a wafer for inspection as described in Bhaskar paragraphs 80-81. Each unit image of the plurality of unit images (one of the image of the multiple images) to include one unit pattern of the unit patterns (structure corresponding to images 46) are acquired at multiple positions which constitute to splitting the full image into acquiring a plurality of unit images (multiple images 46). They are based on respective positions because they are acquired from multiple positions from a respective adjacent identical cell or from a respective neighboring cell).
Claim Objections
Claims 4-5, 8, and 18 are objected to because of the following informalities:
The term “cropping” in claims 4, 8, and 18 lacks antecedent basis.
Claim 8 is inconsistent and contradicts independent claim 1. Claim 1 requires that calculating the respective center positions is based on the blurring. The applicant amended claim 8 and now recites that the calculating the respective center positions is based on the unit images (previously “blurred unit images”), rendering the scope of claim 8 unclear whether the unit images are blurred/non blurred and inconsistent with claim 1. For the purposes of this examination, the examiner assumes that claim 8 requires the unit images to be “blurred unit images” consistent with claim 1.
The phrase “wherein the splitting the full image” should either read: “wherein splitting the full image”.
Appropriate correction is required.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, and 6-19 are rejected under 35 U.S.C. 103 as being unpatentable over Bhaskar et al. (US 20090041332 A1, hereinafter “Bhaskar”) in view of Chen (Chen, H.-C. (2020). Automated Detection and Classification of Defective and Abnormal Dies in Wafer Images. Applied Sciences, 10(10), 3423. https://doi.org/10.3390/app10103423, hereinafter “Chen”) and Nolan et al. (US 20070189628 A1, hereinafter “Nolan”).
Regarding claim 13,
Bhaskar teaches:
An image processing system (Bhaskar: ¶3, “. . .inspection system. . .”; ¶123, “. . .computer system. . .”) comprising:
a memory storing a program therein; and a processor (Bhaskar: ¶123, “. . . computer system 36 . . . having one or more processors, which executes instructions from a memory medium. . .),
wherein, when the program is executed by the processor, the program is configured to:
generate a full image of an area of interest (Bhaskar: ¶80, “. . . acquiring output of an inspection system for a centrally located die on a wafer and one or more dies located on the wafer . . .”; ¶81, “. . . output of the inspection system may be acquired for the entire wafer . . .”; NOTE: The image of an entire wafer is the full image);
in which unit patterns are repeatedly arranged (Bhaskar: ¶80, “ . . . as shown in FIG. 3, wafer 10 includes array 12 of dies formed on the wafer. . .”; NOTE: each die of the array 12 of dies are the unit patterns repeatedly arranged);
set respective reference positions on each of the unit patterns (Bhaskar: ¶83, “. . . The output acquired at the same within die position for multiple dies may be identified by aligning the output acquired for the multiple dies to each other . . . output acquired at the same within die position for multiple dies may be identified by determining the within die position of the output (e.g., using some common reference or coordinate system). Such alignment or determining the within die position may be performed as described further herein . . .”; ¶109, “. . . determining a position of the standard reference die and a position of the output for the one or more test dies with respect to design data space prior to the comparing step and aligning the standard reference die and the output for the one or more test dies based on the positions of the standard reference die and the output for the one or more test dies . . .”; NOTE: The determined position is the respective reference position and used to align dies, the dies being the unit patterns.);
split the full image into a plurality of unit images based on the respective reference positions for each unit image of the plurality of unit images to include one unit pattern of the unit patterns (Bhaskar: Fig. 8-9, ¶130-131, “The multiple images are acquired at multiple positions on the wafer at which the structure is formed. In one embodiment, the multiple positions include positions of the structure in cells having identical designs. . . multiple dies from each focus and exposure point can be averaged together. . . individual structures may have random variations. . . by averaging multiple images of the structure together, either from identical cells (e.g., adjacent identical cells) or from neighboring dies, a composite image can be constructed with much lower LER or random variation than any individual structure. . .”; NOTE: The full image is the output of the inspection system which is an image of a wafer for inspection as described in Bhaskar paragraphs 80-81. Each unit image of the plurality of unit images (one of the image of the multiple images) to include one unit pattern of the unit patterns (structure) are acquired at multiple positions which constitute to splitting the full image into acquiring a plurality of unit images (multiple images). They are based on respective positions because they are acquired from multiple positions from a respective adjacent identical cell or from a respective neighboring cell).; and
merge the plurality of unit images based on the respective reference positions to generate an averaged image (Bhaskar: ¶12, “. . . The method also includes combining the output for the centrally located die and the one or more dies based on within die positions of the output . . .”; ¶84, “. . . determining the within die position of the output (e.g., using some common reference or coordinate system). . .”; ¶131, “. . . the combining step is performed. . . by averaging multiple images of the structure together . . . a composite image can be constructed . . .”; ¶135, “. . . the composite averaged image of the structure shown in FIG. 9 . . .”; NOTE: The “within die positions” of each die are the respective reference positions, which are determined using coordinate system or common reference, and used by Bhaskar for averaging the images of the dies.).
However, Bhaskar fails to teach: blur each of the unit patterns (NOTE: Bhaskar does not explicitly disclose if the smoothing (Bhaskar: para 88 ) and low pass filter (Bhaskar: para 70) are used to blur each of the cropped dies or the full image. Applying smoothing or an optical low pass filtering produces a blurred image); Bhaskar also fails to teach: calculate respective center positions of each of the unit patterns based on the blurring; setting respective reference positions on each of the unit patterns based on the center positions. (NOTE: Bhaskar’s alignment only describes determining “within die positions” and does not explicitly disclose if the “within die position” is a center position identified. Bhaskar only discloses that the position to base the alignment and averaging can be determined using a common reference or coordinate system as described in paragraphs 12, 84, 130-132, and 135. The center position is only one of the coordinates within a die image.)
The analogous art Chen teaches:
blur each of the unit patterns (Chen: pages 6-7, section 2.1.2. step 1, “. . . regularly repeated regions inside the initial template (as shown in Figure 3) . . . The initial template is first smoothed using a two-dimensional Gaussian filter . . .”; NOTE: Applying a Gaussian filter to an image produces a blurred image. Chen’s initial template contains repeated regions, which are the unit patterns. Since the initial template is an image containing the unit patterns, applying a Gaussian filter to the initial template inherently blurs the unit patterns within the image.)
calculate respective center positions of each of the unit patterns based on the blurring (Chen: pages 6-7, section 2.1.2. step 1, “. . . regularly repeated regions inside the initial template (as shown in Figure 3) . . . The initial template is first smoothed using a two-dimensional Gaussian filter . . .”; Chen: page 7, section 2.1.2. steps 2-3, “Thresholding is applied to the filtered template . . . After labeling all bright objects . . . centroid (xL, yL) is recorded . . .”; Chen: page 6: Step 9, “The connected component method is applied to label all bright objects in map RB, and then calculate the centroid of every object”; NOTE: The labeled bright objects are the unit patterns. An image smoothed using a two-dimensional Gaussian filter produces a blurred image. The centroid calculation is based on the blurring because the thresholding is applied to the “filtered template” which applied a Gaussian filter described in Chen 2.1.2 step 1 then a connected component method applied to calculate the centroid of every object.);
It would have been obvious to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention to combine Bhaskar and Chen and implement Chen’s method of: blurring each of the unit patterns; calculating respective center positions of each of the unit patterns based on the blurring and apply the centroid calculation of Chen to Bhaskar’s cropped images 46.
The reason for doing so predict the locations of pattern candidates that possibly contain certain predefined patterns from the detected die patterns and their spatial properties (Chen page 3, 2nd bullet.).
However, the combination of Bhaskar and Chen still fails to teach: set respective reference positions on each of the unit patterns based on the center positions. (NOTE: Bhaskar’s alignment only describes determining “within die positions” as the reference position for averaging and does not explicitly disclose if the “within die position” is a center position. Bhaskar only discloses that the position to base the alignment and averaging can be determined using a common reference or coordinate system as described in paragraphs 12, 83-84, 130-132, and 135. Also the center position is only one of the coordinates within a die image.).
The analogous art Nolan teaches:
set respective reference positions on each of the unit patterns based on the center positions (Nolan: ¶47, “. . . multiple instances of the same glyph image are centroid aligned.”; NOTE: The multiple instances of the same glyph image is the unit patterns. They are centroid aligned, therefore, the respective reference positions used of alignment is based on the centroid which is the center positions.).
It would have been obvious to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention to combine Bhaskar, Chen, and Nolan to perform: setting respective reference positions on each of the unit patterns based on the center positions.
The reason for doing so is to generate a “new universal document imaging format” that “provides the low-production cost and reliable fidelity of image-based formats” (Nolan: ¶28). Additionally, because the “images are precisely aligned and overlaid so that the entire cluster is arranged in one large stack” (Nolan: ¶48).
It would also have been an obvious design choice to perform: setting respective reference positions on each of the unit patterns based on the center positions to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention (NOTE: a center position is a coordinate “within die position” among finite pixel coordinates within an image).
The reason for doing so is to improve Bhaskar’s alignment of the cropped images 46 to easily visualize differences in shapes of the dies.
Regarding claim 14, depending on 13,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing system of claim 13, wherein when the program is executed by the processor, the program is further configured to:
subtract data of the averaged image from data of the full image; and determine whether the unit patterns are defective, based on the subtraction result (Bhaskar: ¶76, “Currently used repeater defect detection approaches for BE typically utilize basically A-B, B-C comparison techniques. In the typical case, A and C are used as the reference dies, and B is used as the test or candidate die in which defects are being detected. . . the reference dies A and C are actually the average or median values of a set of about five dies to about eight dies . . .”; NOTE: A-B is subtraction of data, A and C are the reference dies which is the averaged image of multiple dies. B is the die data of the wafer image, which is the unit pattern data of the full image.);
Regarding claim 15, depending on 13,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing system of claim 13, wherein when the program is executed by the processor, the program is further configured to
calculate a distribution of edge profiles of the unit patterns based on the averaged image (Bhaskar: ¶131, “. . . the combining step is performed such that the composite image has less line edge roughness (LER)) . . .”; ¶135, “. . . the composite averaged image of the structure shown in FIG. 9 shows how the LER is reduced . . .”; ¶136, “. . . detection of systematic defects on a wafer . . . when the defect size is comparable to LER . . .”; NOTE: The LER of the averaged image is the distribution of edge profiles of the unit patterns. It is visually shown in Fig. 9, that the unit patterns with their own LER is averaged to create an average composite image 48 of Fig. 9. The averaged image 48 appears to have straighter edges, thus the reduced LER. Paragraph 135 teaches that the LER can be compared to detect defects on a wafer, therefore, the LER is inherently calculated and must have a value to be used for comparison purposes.)
Regarding claim 16, depending on 13,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing system of claim 13, wherein when the program is executed by the processor, the program is further configured to
measure a width of the unit patterns based on the averaged image (Bhaskar: ¶135, “. . . the defects include systematic defects. Systematic defects occur repeatedly within either an array structure of a single die or within the same structure on multiple die . . . the composite averaged image of the structure shown in FIG. 9 shows how the LER is reduced to allow better sensitivity to systematic line width (LW) variation.”; NOTE: Defects are detected from single die. The system evaluates line width variation. The system knows the variation of the line width, in order for the system to know the variation of the line width, the system inherently measures the width of each of the unit pattern to be used in detection of defects).
Regarding claim 17, depending on 13,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing system of claim 13, wherein when the program is executed by the processor, the program is further configured to
calculate a profile of an image signal of the unit patterns based on the averaged image (Bhaskar: ¶76, “. . . A-B, B-C comparison techniques . . . A and C are used as the reference dies, and B is used as the test or candidate die in which defects are being detected. This type of inspection approach attempts to maximize the ability to capture single isolated signals that are seen as distinct events . . . may be performed using the MDAT algorithm. . . the reference dies A and C are actually the average or median values of a set of about five dies to about eight dies . . . Since BF systems tend to have an RTA subsystem that delivers sub-pixel accuracy. . . “; ¶73, “. . . traditional near square wave edge as shown in FIG. 1(a) could produce the responses shown in FIG. 1 for BF imaging. . .”; NOTE: The captured single isolated signal is a profile of an image signal of a unit pattern which is calculated using MDAT algorithm. It is based on the averaged image because the type of inspection used is the A-B, B-C comparison techniques. A and C are the averaged image, and B is the unit pattern or candidate die. Also, figure 1 shows profile of image signal for BF imaging.).
Regarding claim 18, depending on 13,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing system of claim 13,
wherein each of the plurality of unit images includes a respective one of the unit patterns (Bhaskar: Fig, 8, images 46)
Although Bhaskar teaches smoothing (NOTE: Smoothing results in a blurred image) in relation to alignment errors as described in paragraph 88, Bhaskar is not explicit whether the blurring is done after cropping, Bhaskar fails to teach: wherein cropping the full image into the plurality of unit images is performed before blurring each of the unit patterns.
It would also have been an obvious design choice to include: wherein cropping the full image into the plurality of unit images is performed before blurring each of the unit patterns among finite solutions (cropping before blurring, blurring before cropping) to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention.
The reason for doing so is for parallel processing of smaller cropped images simultaneously making image processing faster. Additionally, blurring before or after cropping results in a same predictable result, a blurred image.
Although Bhaskar teaches smoothing (NOTE: Smoothing results in a blurred image included in Bhaskar’s image processing pipeline) in relation to alignment errors as described in paragraph 88, Bhaskar is not explicit whether the blurring is applied to the cropped images 46, Bhaskar fails to teach: wherein blurring each of the unit patterns includes blurring a respective one of the plurality of unit images, respectively.
It would also have been an obvious design choice to include: wherein blurring each of the unit patterns includes blurring a respective one of the plurality of unit images among finite solutions (blurring a few cropped images respectively, blurring all cropped images respectively) to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention.
The reason for doing so is to ensure uniform and consistent image processing when generating Bhaskar’s averaged composite image.
Although the above combination teaches calculating the respective center positions of each of the unit patterns (Nolan: ¶47, centroid alignment, also see rejection of claim 1.), the above combination still fails to teach: wherein calculating the respective center positions of each of the unit patterns is based on a respective one of the plurality of unit images subjected to blurring (NOTE: Nolan does not teach if the plurality of unit images (glyph images) may be a blurred image.).
It would have been obvious an obvious design choice to choose: wherein calculating the respective center positions of each of the unit patterns is based on a respective one of the plurality of unit images subjected to blurring among finite solutions (based on a blurred image, based on a non-blurred image) to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention.
The reason for doing so is to improve Nolan’s centroid alignment logic. Blurring the images reduces noise of an image and mitigates aliasing error as described by Bhaskar (Bhaskar: ¶70, low pass filtering; ¶88, smoothing; NOTE: Both results in a blurred image), efficiently improving the accuracy of the identification of center of mass or centroid of an image.
Regarding claim 19, depending on 13,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing system of claim 13,
Although Bhaskar teaches smoothing and applying low pass filters (Bhaskar paragraphs 70, 88) which both results in a blurred image, Bhaskar fails to explicitly teach: wherein blurring the unit patterns includes blurring the full image.
The analogous art Chen teaches:
wherein blurring the unit patterns includes blurring the full image (Chen: pages 6-7, section 2.1.2. step 1, “. . . regularly repeated regions inside the initial template (as shown in Figure 3) . . . The initial template is first smoothed using a two-dimensional Gaussian filter . . .”; NOTE: Applying a Gaussian filter to an image produces a blurred image. The full image is the initial template. Chen’s initial template contains repeated regions, which are the unit patterns. Since the initial template is an image containing the unit patterns, applying a Gaussian filter to the initial template inherently blurs the unit patterns within the image.).
wherein calculating the respective center positions of each of the unit patterns is based on the full image processed to be blurred (Chen: pages 6-7, section 2.1.2. step 1, “. . . regularly repeated regions inside the initial template (as shown in Figure 3) . . . The initial template is first smoothed using a two-dimensional Gaussian filter . . .”; Chen: page 7, section 2.1.2. steps 2-3, “Thresholding is applied to the filtered template . . . After labeling all bright objects . . . centroid (xL, yL) is recorded . . .”; Chen: page 6: Step 9, “The connected component method is applied to label all bright objects in map RB, and then calculate the centroid of every object”; NOTE: The initial template is the full image which contains repeated regions. The labeled bright objects are the unit patterns. An image smoothed using a two-dimensional Gaussian filter produces a blurred image. The centroid calculation is based on the blurring because the thresholding is applied to the “filtered template” which applied a Gaussian filter described in Chen 2.1.2 step 1 then a connected component method applied to calculate the centroid of every object.);
It would have been obvious to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention to combine Bhaskar and Chen and include: wherein blurring the unit patterns includes blurring the full image; and wherein calculating the respective center positions of each of the unit patterns is based on the full image processed to be blurred.
The reason for doing so predict the locations of pattern candidates that possibly contain certain predefined patterns from the detected die patterns and their spatial properties (Chen page 3, 2nd bullet.).
Regarding claims 1, and 8-9, method claims 1, and 8-9 respectively are drawn to the methods corresponding to the program of using same as claimed in apparatus claim 13, and 18-19 respectively. Therefore, method claims 1, and 8-9 respectively correspond to the program in the apparatus of claims 13, and 18-19 respectively, and are rejected for the same reasons of obviousness as used above.
Regarding claim 6, depending on 1,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing method of claim 1,
wherein each of the respective center positions is substantially coincident with a respective one of the respective reference positions (Bhaskar: ¶46, “FIG. 9 is a schematic diagram illustrating one example of a composite image of the structure shown in the images of FIG. 8 generated by combining the multiple images of the structure shown in FIG. 8”; NOTE: It can be observed from Bhaskar Fig. 9 that the dies images 46 center positions are overlapping. Figure 9 illustrates alignment of multiple dies with overlapping center positions. Therefore, the alignment coordinates that Bhaskar used is substantially coincident with a respective one of the reference positions. Additionally, as discussed in the rejection of claim 1, Nolan uses the centroids to align glyph images, therefore, the centroids or center positions are used as the reference positions for alignment, not just substantially coincident.).
Regarding claim 7, depending on 1,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing method of claim 1,
wherein setting each of the respective reference positions includes correcting a respective center position by a preset offset (Nolan: ¶12, “. . . aligning each glyph image involves performing a centroid alignment, wherein the centroid is the center of mass of the glyph image, or performing an offset-centroid alignment, wherein an offset is added to the centroid to determine the alignment of the glyph image.”; NOTE: The setting each of the reference position is the determination of alignment of the glyph image. The correcting a respective center position by a preset offset is the adding an offset to the centroid. The centroid is the respective center position. The offset is preset because it is needed to determine the alignment of the glyph image. When the offset is added to the centroid, it corrects a respective center position.)
Regarding claim 10, depending on 1,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing method of claim 1, wherein blurring each of the unit patterns includes performing a Gaussian blurring process thereon (Chen: pages 6-7, section 2.1.2. step 1, “. . . regularly repeated regions inside the initial template (as shown in Figure 3) . . . The initial template is first smoothed using a two-dimensional Gaussian filter . . .”; NOTE: performing a Gaussian blurring is equivalent to smoothing using a Gaussian filter).
Regarding claim 11, depending on 1,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing method of claim 1, wherein generating the full image includes imaging the area of interest using a scanning electron microscope (SEM) (Bhaskar: ¶57, “. . . For example, a short loop review inspection cycle optimization (RICO) type experiment may be performed between the wafer inspection system and the defect review and/or inspection system (e.g., a scanning electron microscope (SEM).”).
Regarding claim 12, depending on 1,
The combination of Bhaskar, Chen and Nolan teaches:
The image processing method of claim 1, wherein calculating the respective center positions of each of the unit patterns includes calculating a position where gray-level data of each unit pattern has a substantially maximum value (Chen: page 4, Step 2, “The original image is converted into a grayscale image”; Chen: page 6, step 9, “The connected component method is applied to label all bright objects in map RB, and then calculate the centroid of every object. . .”; Chen: page 6, step 12, “the centroid of every local peak is found . . . then localized. NOTE: A grayscale image includes gray-level data or intensity values. The centroid of every local peak is position where gray-level data of each unit pattern has a substantially maximum value. Chen’s system distinguishes bright objects, bright objects has substantially maximum value of gray-level, the brighter the object, the higher the gray intensity level.).
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Bhaskar et al. in view of Chen and Nolan further in view of Sah et al. (US 20140355873 A1, hereinafter “Sah”)
Regarding claim 2, depending on claim 1,
The combination of Bhaskar, Chen, and Nolan teaches:
The image processing method of claim 1, further comprising
However, the combination of Bhaskar, Chen and Nolan fails to teach: calculating a period at which the unit patterns are repeatedly arranged, based on the full image.
The analogous art Sah teaches:
calculating a period at which the unit patterns are repeatedly arranged, based on the full image (Sah: ¶36, “FIG. 1 . . . and a repetition cycle calculation unit 15 which calculates distances between the unit patterns in each effective layer and calculates a pattern repetition cycle . . . The repetitive pattern detection apparatus receives an input image 11 and calculates a repetition cycle 16 . . .”; NOTE: The period calculated is the calculated pattern repetition cycle).
It would have been obvious to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention to combine Bhaskar, Chen, Nolan and Sah and include: calculating a period at which the unit patterns are repeatedly arranged, based on the full image.
The reason for doing so is for “detecting a defect in a test image using a repetitive pattern detected in an input image in order to detect a defect by visual detection without a reference image” (Sah: ¶10).
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Bhaskar et al. in view of Chen and Nolan further in view of Sah further in view of Morioka et al. (US 20010021015 A1, hereinafter “Morioka”).
Regarding claim 3, depending on 2,
The combination of Bhaskar, Chen, Nolan and Sah teaches:
The image processing method of claim 2,
However, the above combination fails to teach: wherein calculating the period includes performing a fast Fourier transform (FFT) on an image signal of the full image. (NOTE: Sah calculates the period using median value of distance values between unit patterns to calculate a correct repetition cycle as described in paragraph 51.)
The analogous art Morioka teaches:
wherein calculating the period includes performing a fast Fourier transform (FFT) on an image signal of the full image (Morioka: ¶139, “In the pitch detecting means 2212, a pattern repetition pitch . . . are measured from a detected signal . . . ¶144, “In the pitch detecting . . . a Fourier transform processing for detected image is performed by the FFT circuit 2242 . . .”; ¶160, “When the wafer 1001 has been conveyed to a position in which the detecting optical system 2101 can take in repetitive patterns on the wafer, a signal detected by the detector 2107 is analyzed for frequency by the FFT circuit . . . to provide a spatial frequency which is a peak in the frequency region. . . For this frequency analysis, fast Fourier transform is most desirable from the standpoint of processing speed . . .; NOTE: The period is the pattern repetition pitch. The image signal is the signal detected by the detector 2107. Morioka’s system takes in an image of a wafer with repetitive patterns, the process to measure the pattern repetition pitch includes using fast Fourier transform FFT. The wafer 1001 is the full image because it is used with an optical system 1001.).
It would have been obvious to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention to combine Bhaskar, Chen, Nolan, Sah and Morioka and include: wherein calculating the period includes performing a fast Fourier transform (FFT) on an image signal of the full image.
The reason for doing so is because for “frequency analysis, fast Fourier transform is most desirable from the standpoint of processing speed” (Morioka: ¶160).
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Bhaskar et al. in view of Chen and Nolan further in view of Sah further in view of Ghosh et al. (US 20050281462 A1, hereinafter “Ghosh”)
Regarding claim 4, depending on 2,
The combination of Bhaskar, Chen, Nolan and Sah teaches:
The image processing method of claim 2,
Bhaskar also teaches cropping the full image into the plurality of unit images, which is the images 46 as discussed in claim 1 rejection.
However, the above combination fails to teach: wherein cropping the full image into the plurality of unit images is based on the period.
The analogous art Ghosh teaches:
wherein cropping the full image into the plurality of unit images is based on the period (Ghosh: ¶112, “. . . example of baseline processing involves filtering sources of illumination which have a period (width) . . .”; ¶4, “. . . automatically cropping the image containing multiple microarray images to form a group of single images, each containing only one microarray image cropped from the image containing multiple microarray images . . .”; ¶72, “After inputting the multipack image . . . perform the cropping operations at step 206. The system preferably uses a projection-based algorithm to locate the features on the microarrays on the multipack image. . .”; ¶96, “. . . the system . . . find the spacing between peaks. The spaces between each pair of adjacent peaks are calculated and tabulated, after which, the median difference between adjacent peaks is calculated. The median value is then set to be the feature spacing, i.e. distance between adjacent features in the dimension being considered . . .”; ¶97, “. . . The peak spacing may be further used to determine group spacing, e.g., distance between microarrays 110 . . .”; ¶73, “. . . Once the locations of the microarrays 110 have been determined on the multipack image 100, the system crops the images at step 208, thereby creating a single image for each microarray . . .”; NOTE: the plurality of unit images are the cropped images each containing only one microarray image. The full image is the image containing multiple microarray image or the multipack image. The spacing between peaks or peak spacing is the period. Therefore, Ghosh’s cropping is based on the period because Ghosh’s system determines the feature spacing first before cropping.).
It would have been obvious to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention to combine Bhaskar, Chen, Nolan, Sah and Ghosh to include: wherein cropping the full image into the plurality of unit images is based on the period.
The reason for doing so is for “automatically cropping the image containing multiple microarray images to form a group of single images, each containing only one microarray image cropped from the image containing multiple microarray images” (Ghosh: ¶4).
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Bhaskar et al. in view of Chen and Nolan further in view of Ghosh.
Regarding claim 5, depending on 1,
The combination of Bhaskar, Chen, and Nolan teaches:
The image processing method of claim 1,
However, the above combination fails to teach: wherein cropping the full image into the plurality of unit images is based on a preset size.
The analogous art Ghosh teaches:
wherein the splitting the full image into the plurality of unit images is based on a preset size (Ghosh: ¶73, “. . . the microarrays 110 have been determined on the multipack image 100, the system crops the images at step 208, thereby creating a single image for each microarray 110 . . .”; Ghosh: ¶67, “. . . the user may . . .specify cropping parameters, such as layout characteristics of the multipack array at step 204 . . .; ¶70, “. . . the system will default to establish margins of predetermined default size . . . accurately locating and cropping the microarrays . . .”; NOTE: The full image is the multipack image, the plurality of unit images are the single images for each microarray. specifying cropping parameters is the “preset size” because it is set prior to splitting or cropping. Also, a predetermined default size is a preset size.)
It would have been obvious to a person having ordinary skill in the art (PHOSITA) before the effective filing date of the claimed invention to combine Bhaskar, Chen, Nolan and Ghosh to include: wherein the splitting the full image into the plurality of unit images is based on a preset size.
The reason for doing so is because “this increases the probabilities of accurately locating and cropping the microarrays, particularly when microarrays having skewed features/probes or other errors are present, such that the projections are not very well defined” (Ghosh: ¶70).
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PATRICK GALERA whose telephone number is (571)272-5070. The examiner can normally be reached Mon-Fri 0800-1700 ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, King Poon can be reached at 571-270-0728. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/PATRICK P GALERA/Examiner, Art Unit 2617
/KING Y POON/Supervisory Patent Examiner, Art Unit 2617
Prosecution Insights