A chip defect area determination method, device, equipment and storage medium
Through automated image processing technology, including arrangement direction detection, segmentation and cropping, the accurate determination of chip defect areas is achieved, solving the problems of low efficiency and high misjudgment rate of traditional methods, and improving detection efficiency and accuracy.
Patent Information
- Application Number
- CN202411245302.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-06
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-06
AI Technical Summary
Traditional chip defect positioning methods rely on manual scanning, which are inefficient and easily lead to fatigue misjudgment, and cannot meet detection needs.
By acquiring the original chip image, performing arrangement direction detection and segmentation processing, generating segmented projected images, counting waveform extreme points, estimating period length, cropping images, and comparing feature indicators, and automatically determining defect locations.
No manual participation is required, reducing costs, improving detection efficiency, and improving the accuracy and detection efficiency of defective areas.
Smart Images

Figure CN118762014B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of semiconductor detection technology, and in particular to a chip defect area determination method, device, equipment and storage medium. Background Art
[0002] Chip failure analysis is a complex and technology-intensive process used to determine the failure modes and root causes of semiconductor devices, integrated circuits or chips. This process is critical to improving product quality, improving design and manufacturing processes, and ensuring product reliability.
[0003] In the traditional process of determining and locating defect locations, the method commonly used in the semiconductor field is for inspectors to scan each layer of the sample to be inspected, bank by bank, instance by instance, and array by visually observing the bit lines (BL) and word lines (WL) to determine whether there are any voltage contrast anomalies to find the detailed defect location.
[0004] In order to observe defects, the imaging equipment needs to have a relatively high magnification, and the area of the sample to be inspected is small at a time, which results in a large workload for inspectors and a long processing time, resulting in low inspection efficiency. In addition, high-density repetitive work can easily cause fatigue, and long-term work can easily lead to misjudgment or missed judgment, resulting in poor inspection accuracy and failure to meet actual needs. Summary of the invention
[0005] The present invention provides a chip defect area determination method, device, equipment and storage medium to accurately determine the chip defect area without manual intervention, thereby reducing labor costs and improving the detection efficiency of determining the defect area.
[0006] According to one aspect of the present invention, a method for determining a chip defect area is provided. The method comprises:
[0007] Obtaining the original chip image corresponding to the target chip;
[0008] Detecting the arrangement direction of the unit structure in the original chip image, determining the image cutting direction corresponding to the original chip image, and segmenting the original chip image according to the image cutting direction to obtain a first segmented image;
[0009] For each of the first segmented images, based on directions perpendicular to each other of the image cutting directions, the first segmented images are subjected to pixel value statistical processing to obtain a segmented projection image, wherein the segmented projection image includes a plurality of waveform extreme value points, and the waveform extreme value points are peak extreme value points or trough extreme value points;
[0010] Determine the estimated period length of the unit structure in the first segmented image according to the segmented projection image, and perform cropping processing on the first segmented image based on the estimated period length to obtain each second segmented image;
[0011] For each of the second segmented images, the current second segmented image is compared with other second segmented images to determine a target second segmented image with defects, and based on the target second segmented image, the defect position of the target chip is determined.
[0012] According to another aspect of the present invention, there is provided a device for determining a chip defect area, the device comprising:
[0013] An original chip image acquisition module is used to acquire the original chip image corresponding to the target chip;
[0014] An original chip image segmentation module is used to detect the arrangement direction of the unit structure in the original chip image, determine the image cutting direction corresponding to the original chip image, and segment the original chip image according to the image cutting direction to obtain a first segmented image;
[0015] a segmented projection image determination module, configured to perform pixel value statistical processing on each of the first segmented images based on directions perpendicular to each other of the image cutting directions, to obtain a segmented projection image, wherein the segmented projection image includes a plurality of waveform extreme value points, and the waveform extreme value points are peak extreme value points or trough extreme value points;
[0016] A first segmented image cropping module, configured to determine an estimated period length of a unit structure in the first segmented image according to the segmented projection image, and to crop the first segmented image based on the estimated period length to obtain respective second segmented images;
[0017] The chip defect position determination module is used to compare the current second segmented image with other second segmented images for each of the second segmented images, determine the target second segmented image with defects, and determine the defect position of the target chip based on the target second segmented image.
[0018] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:
[0019] at least one processor; and
[0020] a memory communicatively connected to the at least one processor; wherein,
[0021] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the chip defect area determination method described in any embodiment of the present invention.
[0022] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the chip defect area determination method described in any embodiment of the present invention when executed.
[0023] The technical solution of the embodiment of the present invention is to obtain the original chip image corresponding to the target chip. The arrangement direction of the unit structure in the original chip image is detected, the image cutting direction of the original chip image is determined, and the original chip image is segmented according to the image cutting direction to obtain a first segmented image. For each of the first segmented images, the first segmented image is statistically processed based on the mutually perpendicular directions of the image cutting direction to obtain a segmented projection image. According to the segmented projection image, the estimated period length of the unit structure in the first segmented image is determined, and based on the estimated period length, the first segmented image is cropped to obtain each second segmented image. By cropping the original chip image step by step into a second segmented image that can be analyzed, and then comparing the feature indicators of the current second segmented image with other second segmented images, the target second segmented image with defects is determined, so that the chip defect area can be accurately determined based on the target second segmented image. This process does not require manual participation, which can reduce labor costs. At the same time, image processing technology can quickly detect whether each second segmented image has defects, thereby improving the detection efficiency of determining defective areas.
[0024] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0026] Figure 1 is a flow chart of a chip defect area determination method provided according to Embodiment 1 of the present invention;
[0027] Figure 2is a schematic diagram of an original chip image provided according to the first embodiment of the present invention;
[0028] Figure 3 is a schematic diagram of a first segmented image provided according to Embodiment 1 of the present invention;
[0029] Figure 4 is a schematic diagram of a segmented projection image provided according to Embodiment 1 of the present invention;
[0030] Figure 5 is a schematic diagram of a projection image of a target second image provided according to Embodiment 1 of the present invention;
[0031] Figure 6 is a schematic diagram of a defect position image of a target chip provided in Embodiment 1 of the present invention;
[0032] Figure 7 is a flow chart of a chip defect area determination method provided according to Embodiment 2 of the present invention;
[0033] Figure 8 is a schematic diagram of a horizontal axis projection image provided according to Embodiment 2 of the present invention;
[0034] Fig. 9 is a structural diagram of a chip defect area determination device provided according to Embodiment 3 of the present invention;
[0035] Fig.10 It is a schematic diagram of the structure of an electronic device for implementing the chip defect area determination method according to an embodiment of the present invention. DETAILED DESCRIPTION
[0036] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0037] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0038] Embodiment 1
[0039] Figure 1 A flowchart of a chip defect area determination method is provided for the first embodiment of the present invention. This embodiment is applicable to the case where a chip defect area is automatically detected. The method can be performed by a chip defect area determination device. The chip defect area determination device can be implemented in the form of hardware and / or software. The chip defect area determination device can be configured in an electronic device. Figure 1 As shown, the method includes:
[0040] S101, obtaining an original chip image corresponding to a target chip.
[0041] The target chip may be a chip to be defect-detected. The target chip includes a plurality of minimum unit structures. In the technical solution of the present invention, the target chip includes at least three minimum unit structures to accurately determine the defect area. The original chip image may refer to the original scanned image of the target chip image by the scanning device. For example, the original chip image may be a grayscale image or a color image.
[0042] Specifically, a series of high-resolution original scanned images are obtained through an image scanning device (e.g., a scanning electron microscope), ensuring that the images cover the key areas of the chip and have sufficient magnification to clearly display the smallest unit structure. The original scanned images are subjected to noise removal and image enhancement processing to obtain the original chip image.
[0043] In the disclosed embodiment, the specific imaging magnification of the original chip image varies according to the manufacturing process, and is applicable to images containing more than three complete periodic minimum unit structures in each image to determine the potential contrast abnormality defect location. Figure 2 is a schematic diagram of the original chip image provided by the present disclosure, such as Figure 2As shown in the figure, taking the original chip image as a grayscale image as an example, the black part in the figure is the background image, and the white dots are the minimum unit structure with periodicity. It is worth noting that black and white do not represent the corresponding structure in a fixed manner. In other original chip image schematics, the pixel values may be reversed, that is, black can be represented as the minimum unit structure with periodicity, and white can be represented as the background image.
[0044] S102, detecting the arrangement direction of the unit structure in the original chip image, determining the image cutting direction corresponding to the original chip image, and segmenting the original chip image according to the image cutting direction to obtain a first segmented image.
[0045] It should be noted that, in consideration of process, cost and performance, high-density integrated standardized array design is often used in chip design, so the minimum unit structure in the target chip has a periodic arrangement characteristic. Segmenting the original chip image along the periodic arrangement direction can improve the detection efficiency and accuracy of the defect location.
[0046] The image cutting direction may refer to the direction in which the original chip image is cut, which is consistent with the arrangement direction of the minimum unit structure in the target chip. Figure 3 is a schematic diagram of a first segmented image provided by an embodiment of the present disclosure. Figure 3 As shown, the first segmented image may refer to a segmented image of the original chip image. The number of the first segmented image is at least one.
[0047] Specifically, the periodicity of the arrangement direction of the unit structure in the original chip image is detected to determine whether the arrangement periodicity direction of the unit structure is a horizontal axis period or a vertical axis period, and then the arrangement periodicity direction is used as the image cutting direction, and the original chip image is segmented along the image cutting direction, thereby retaining the periodicity of the first segmented image, thereby improving the detection efficiency and accuracy of the defect position.
[0048] S103 . For each of the first segmented images, based on directions perpendicular to each other of the image cutting directions, perform pixel value statistics processing on the first segmented image to obtain a segmented projection image.
[0049] The segmented projection image may be a projection image obtained by performing pixel value statistics processing on the first segmented image. Figure 4 Schematic diagram of the segmented projection image provided by the embodiment of the present disclosure. Figure 4 As shown, the segmented projection image includes at least one waveform extreme point, which can be a peak extreme point or a trough extreme point. The peak is the point with the highest cumulative value of pixel value statistics, and the trough is the point with the lowest cumulative value of the accumulated pixels.
[0050] It should be noted that pixel value statistical processing is an image processing technology that mainly involves operations such as accumulation and smoothing of pixel values in an image. Among them, pixels are the basic components of an image. Pixel values can contain color information (such as RGB values) or grayscale values. Accumulation is the process of adding a group of pixels (actually pixel values). Pixel value statistical processing can be applied to a variety of image processing scenarios, including but not limited to image analysis, image enhancement, and image fusion.
[0051] Specifically, the first segmented image is subjected to mean filtering to reduce noise interference. The image cutting direction of the first segmented image is determined, and the first segmented image is subjected to pixel value statistical processing along the mutually perpendicular directions of the image cutting directions, so as to obtain a segmented projection image. For example, Figure 3 As shown, the image cutting direction is horizontal, then the pixel values of the first segmented image are statistically processed in the vertical direction to obtain a projection array in the horizontal direction, and the projection array is smoothed by moving average, so that the segmented projection image (such as Figure 4 as shown).
[0052] S104 . Determine an estimated period length of the unit structure in the first segmented image according to the segmented projection image, and crop the first segmented image based on the estimated period length to obtain respective second segmented images.
[0053] The estimated period length may refer to the estimated length occupied by the minimum unit structure in the first segmented image. It should be noted that the estimated period length is an estimated value, and the period lengths occupied by the minimum unit structure are not exactly the same. In the segmented projection images corresponding to different areas of the target chip, the estimated period length may include a single minimum unit structure or multiple minimum unit structures. The second image may refer to a segmented image obtained by cropping the first segmented image.
[0054] Specifically, for the distance between each group of adjacent waveform extreme value points in the segmented avatar image, the waveform extreme value point autocorrelation score is calculated, and the distance between adjacent waveform extreme value points with the highest autocorrelation score is used as the estimated cycle length. After determining the estimated cycle length, the estimated cycle length is determined to determine the corresponding cropped image length in the first segmented image, and the first segmented image is cropped and divided into a plurality of second segmented images with reference to the cropped image length.
[0055] Exemplarily, determining the estimated period length of the unit structure in the first segmented projection image based on the segmented projection image includes: determining the median extreme point of all the waveform extreme points in the segmented projection image; constructing a period estimation interval based on the median extreme point and a predetermined preset waveform length; and performing periodic detection processing on the unit structure of the period estimation interval to determine the estimated period length.
[0056] Among them, the median extreme point may refer to the median point of all the waveform extreme points in the segmented projection image. The preset waveform length may be set according to actual conditions, may be determined according to historical waveform length data, or may be set according to experience. In the disclosed embodiment, the preset waveform length may be one-third of the length of the complete waveform in the segmented projection image. The period estimation interval may refer to the interval length of the segmented projection image of the preset waveform length. It should be noted that the period estimation interval may take the median extreme point as the center point and cut off the waveform length of the preset waveform length at both ends, or may take the median extreme point as any endpoint of the period estimation interval and cut off the waveform length of the preset waveform length.
[0057] Specifically, all waveform extreme value points in the segmented projection image are arranged in order according to the size of the extreme value points, and the median extreme value point is determined. A plurality of cycle estimation intervals are obtained in the segmented projection image with a predetermined preset waveform length combined with the median extreme value point spacing. The autocorrelation score of the cycle estimation interval is calculated. The cycle distance with the highest autocorrelation score is determined as the estimated cycle length.
[0058] Exemplarily, the cropping of the first segmented image based on the estimated cycle length includes: determining an image boundary of the first segmented image; and cropping the first segmented image based on the estimated cycle length and the image boundary.
[0059] It should be noted that when the image boundary in the technical solution of the embodiment of the present disclosure is uncertain, it is easy to be disturbed by noise points outside the boundary. Therefore, it is necessary to determine the image boundary so that the second segmented image can be obtained more accurately, thereby improving the accuracy of determining the defect area.
[0060] Specifically, the segmented projection image of the first segmented image can be intercepted in two opposite directions using the estimated cycle length based on the median extreme point until there is no waveform extreme point in the intercepted image, and the last intercepted image is the image boundary of the first segmented image. Then, according to the estimated cycle length and the image boundary, the first segmented image is cropped into multiple second segmented images to improve the accuracy of determining the defect area.
[0061] S105 . For each of the second segmented images, compare the current second segmented image with other second segmented images to determine a target second segmented image with defects, and determine a defect position of the target chip based on the target second segmented image.
[0062] Specifically, for each second segmented image, an image index of the second segmented image is determined, the image index of the current second segmented image is compared with the other second segmented images, a target second segmented image inconsistent with the other second segmented images is determined, and the target second segmented image is determined to have a defect. According to the image identification information of the target second segmented image, the defect position of the target chip is determined in the original scanned image or the original chip image.
[0063] Exemplarily, comparing the current second segmented image with other second segmented images to determine the target second segmented image with defects includes: determining a first evaluation index of the current second segmented image and second evaluation indexes of the other second segmented images. Comparing the first evaluation index with each of the second evaluation indexes to determine the target second segmented image with defects.
[0064] The evaluation index may refer to image evaluation information of the second segmented image. The first evaluation index may refer to the evaluation index of the current second segmented image, and the second evaluation index may refer to the collective name of the evaluation indexes of other second segmented images. Exemplarily, the first evaluation index and the second evaluation index at least include peak and trough values, cross-correlation coefficients with other second segmented images, and waveform similarity coefficients.
[0065] Specifically, calculate whether there are outliers in the peaks and troughs of the current second segmented image, calculate the cross-correlation coefficients with other second segmented waveforms, and calculate the waveform similarity coefficient distances with other second segmented images. For the evaluation indicators obtained above, use the OPTICS algorithm for density clustering to obtain the reachable distances between the clusters where the defect is located and the clusters where the pattern components are located. The above features at the location of the defect are usually an outlier that does not belong to any cluster, and the distances between its various indicators and the indicators of adjacent samples will also appear as abnormalities, which can be seen as abnormalities in the reachable distances in the clustering results. Alternatively, by setting a similarity threshold, all similarities beyond a certain degree from the cluster center can be marked as abnormal.
[0066] Figure 5 Schematic diagram of the projection image corresponding to the target second image. Figure 6 Figure 1 is a schematic diagram of the defect location image of the target chip. Figure 5 and 6 As shown, the defect position of the target chip can be obtained by combining the above evaluation index characteristics with the position of the projection image corresponding to the target second image.
[0067] The technical solution of the embodiment of the present invention is to obtain the original chip image corresponding to the target chip. The arrangement direction of the unit structure in the original chip image is detected, the image cutting direction of the original chip image is determined, and the original chip image is segmented according to the image cutting direction to obtain a first segmented image. For each of the first segmented images, the first segmented image is statistically processed based on the mutually perpendicular directions of the image cutting direction to obtain a segmented projection image. According to the segmented projection image, the estimated period length of the unit structure in the first segmented image is determined, and based on the estimated period length, the first segmented image is cropped to obtain each second segmented image. By cropping the original chip image step by step into a second segmented image that can be analyzed, and then comparing the current second segmented image with other second segmented images, the target second segmented image with defects is determined, so that the chip defect area can be accurately determined based on the target second segmented image, without manual participation, which can reduce labor costs. At the same time, through image processing technology, it is possible to quickly detect whether each second segmented image has defects, thereby improving the detection efficiency of determining defective areas.
[0068] Embodiment 2
[0069] Figure 7 This is a flow chart of a chip defect area determination method provided in the second embodiment of the present invention. Based on the above embodiments, this embodiment further refines the determination of the image cutting direction corresponding to the original chip image. Figure 7 As shown, the method includes:
[0070] S201, obtaining an original chip image corresponding to a target chip.
[0071] S202: Perform pixel value statistics processing on the original chip image to obtain an initial projection image.
[0072] The initial projection image may refer to a pixel projection trend diagram of the initial projection image on the horizontal axis or the vertical axis. The initial projection image may be an image obtained by performing pixel value statistics along the horizontal axis and the vertical axis of the original chip image. Exemplarily, the initial projection image includes a horizontal axis projection image and a vertical axis projection image.
[0073] Specifically, the image pixel values are counted along the horizontal axis and the vertical axis of the original chip image, so that the horizontal axis projection image and the vertical axis projection image can be obtained, that is, the initial projection image is obtained.
[0074] S203 , determining an image segmentation region corresponding to the original chip image according to the waveform extreme value points in the initial projection image.
[0075] The image segmentation area may refer to the segmentation area after the original chip image is divided into regions. The image segmentation area may be determined by dividing according to the waveform extreme value points in the initial projection image.
[0076] Specifically, for the transverse projection image, the distribution of the waveform extreme value points in the transverse projection image is analyzed, and based on the positional relationship of the waveform extreme value points, the segmentation area in the transverse projection image is determined, and then the image segmentation area in the original chip image is determined. For the longitudinal projection image, the image segmentation area in the original chip image is determined based on the same technical means.
[0077] Exemplarily, determining the image segmentation region corresponding to the original chip image according to the waveform extreme value points in the transverse projection image includes:
[0078] Performing waveform extreme point detection processing on the transverse axis projection image to determine each waveform extreme point in the transverse axis projection image;
[0079] Isolating each waveform extreme point in the transverse axis projection image to obtain each isolated waveform image;
[0080] For each isolated waveform image, the original chip image is segmented according to the isolated waveform image to obtain an image segmentation area.
[0081] Figure 8 Schematic diagram of a horizontal axis projection image provided by an embodiment of the present disclosure. Figure 8 As shown, it can be Figure 8 The peak points in are taken as the extreme points of each waveform. Figure 8 ,Will Figure 8 Each waveform extreme point in the image is isolated along the longitudinal direction, and each isolated waveform image in the transverse projection image can be determined. According to the isolated waveform image, the segmentation interval of the original chip image is determined, and then the image segmentation area is obtained.
[0082] S204: Determine the regional periodicity scores of the image segmentation regions in directions perpendicular to the segmentation directions, and determine a total value of the image periodicity scores corresponding to the initial projection image according to the regional periodicity scores.
[0083] The total value of the image periodicity score includes the total value of the horizontal axis periodicity score and the total value of the vertical axis periodicity score.
[0084] Specifically, for each image segmentation region segmented along the horizontal axis, its periodicity score on the vertical axis is calculated. Each periodicity score is statistically processed to obtain the total periodicity score on the horizontal axis. For each image segmentation region segmented along the vertical axis, the total periodicity score on the vertical axis is determined based on the same technical means.
[0085] The method of determining the regional periodicity score of the image segmentation area in directions perpendicular to the segmentation directions includes: determining the regional segmentation direction of the image segmentation area, and determining the segmentation directions perpendicular to the regional segmentation direction; performing pixel value statistical processing on the image segmentation area based on the segmentation directions perpendicular to each other to obtain a regional projection image; performing Fourier transform processing on the regional projection image to determine the regional main frequency corresponding to the image segmentation area; and determining the regional periodicity score corresponding to the image segmentation area based on the main frequency amplitude of the regional main frequency.
[0086] Specifically, for the image segmentation area segmented along the horizontal axis, the segmentation direction of the area segmentation is perpendicular to each other as the vertical axis direction, and the pixel value statistics of the image segmentation area are processed on the vertical axis to obtain the regional projection image. The regional projection image is processed by Fourier transform to determine the various frequency components in the regional projection image and the amplitudes corresponding to the various frequency components. The frequency with the largest amplitude is determined as the main frequency, and the ratio of the amplitude of the main frequency component to the total amplitude of all frequency components is determined as the regional periodicity score corresponding to the image segmentation area. Similarly, for the image segmentation area segmented along the vertical axis, the regional periodicity score can be obtained.
[0087] S205: When the total value of the horizontal axis periodicity fraction is greater than or equal to the total value of the vertical axis periodicity fraction, determine that the image cutting direction is horizontal cutting; otherwise, determine that the image cutting direction is vertical cutting.
[0088] Specifically, the total value of the horizontal axis periodicity score and the total value of the vertical axis periodicity score are compared, and when the total value of the horizontal axis periodicity score is greater than or equal to the total value of the vertical axis periodicity score, the image cutting direction is determined to be horizontal cutting. When the total value of the horizontal axis periodicity score is less than the total value of the vertical axis periodicity score, the image cutting direction is determined to be horizontal cutting.
[0089] S206 , segmenting the original chip image according to the image cutting direction to obtain a first segmented image.
[0090] S207 . For each of the first segmented images, based on directions perpendicular to each other of the image cutting directions, perform pixel value statistics processing on the first segmented image to obtain a segmented projection image.
[0091] S208 . Determine an estimated period length of the unit structure in the first segmented image according to the segmented projection image, and crop the first segmented image based on the estimated period length to obtain respective second segmented images.
[0092] S209 , for each of the second segmented images, compare the current second segmented image with other second segmented images to determine a target second segmented image with defects, and determine the defect position of the target chip based on the target second segmented image.
[0093] The technical solution of the embodiment of the present invention obtains an initial projection image by performing pixel value statistical processing on the original chip image. According to the waveform extreme point in the initial projection image, the image segmentation area corresponding to the original chip image is determined. The regional periodicity score of the image segmentation area in the mutually perpendicular direction of the segmentation direction is determined, and the total value of the image periodicity score corresponding to the initial projection image is determined according to the regional periodicity score. When the total value of the horizontal axis periodicity score is greater than or equal to the total value of the vertical axis periodicity score, the image cutting direction is determined to be a horizontal cutting, otherwise, the image cutting direction is determined to be a vertical cutting, and the periodicity of the original chip image after cutting is retained, thereby improving the detection efficiency of determining the defective area. At the same time, through periodicity, the image detection technology is used to accurately determine the chip defect area without manual participation, thereby reducing labor costs.
[0094] Embodiment 3
[0095] Fig. 9 This is a schematic diagram of the structure of a chip defect area determination device provided in Embodiment 3 of the present invention. Fig. 9 As shown, the device comprises:
[0096] The original chip image acquisition module 301 is used to acquire the original chip image corresponding to the target chip;
[0097] The original chip image segmentation module 302 is used to detect the arrangement direction of the unit structure in the original chip image, determine the image cutting direction corresponding to the original chip image, and segment the original chip image according to the image cutting direction to obtain a first segmented image;
[0098] The segmented projection image determination module 303 is used to perform pixel value statistical processing on each of the first segmented images based on the mutually perpendicular directions of the image cutting directions to obtain a segmented projection image, wherein the segmented projection image includes a waveform extreme value point, and the waveform extreme value point is a peak extreme value point or a trough extreme value point;
[0099] A first segmented image cropping module 304 is used to determine the estimated period length of the unit structure in the first segmented image according to the segmented projection image, and to crop the first segmented image based on the estimated period length to obtain each second segmented image;
[0100] The chip defect position determination module 305 is used to compare the current second segmented image with other second segmented images for each of the second segmented images, determine the target second segmented image with defects, and determine the defect position of the target chip based on the target second segmented image.
[0101] The technical solution of the embodiment of the present invention is to obtain the original chip image corresponding to the target chip. The arrangement direction of the unit structure in the original chip image is detected, the image cutting direction of the original chip image is determined, and the original chip image is segmented according to the image cutting direction to obtain a first segmented image. For each of the first segmented images, the first segmented image is statistically processed based on the mutually perpendicular directions of the image cutting direction to obtain a segmented projection image. According to the segmented projection image, the estimated period length of the unit structure in the first segmented image is determined, and based on the estimated period length, the first segmented image is cropped to obtain each second segmented image. By cropping the original chip image step by step into a second segmented image that can be analyzed, and then comparing the current second segmented image with other second segmented images, the target second segmented image with defects is determined, so that the chip defect area can be accurately determined based on the target second segmented image, without manual participation, which can reduce labor costs. At the same time, through image processing technology, it is possible to quickly detect whether each second segmented image has defects, thereby improving the detection efficiency of determining defective areas.
[0102] Optionally, the original chip image segmentation module 302 includes:
[0103] An initial projection image determination unit, used for performing pixel value statistical processing on the original chip image to obtain an initial projection image, wherein the initial projection image includes a horizontal axis projection image and a vertical axis projection image;
[0104] An image segmentation region determination unit, used to determine the image segmentation region corresponding to the original chip image according to the waveform extreme value points in the initial projection image;
[0105] a periodicity total value determination unit, configured to determine a regional periodicity score of the image segmentation region in a direction perpendicular to the segmentation direction, and determine a total image periodicity score value corresponding to the initial projection image according to the regional periodicity score, wherein the total image periodicity score value includes a total horizontal axis periodicity score value and a total vertical axis periodicity score value;
[0106] The image cutting direction determining unit is used to determine that the image cutting direction is horizontal cutting when the total value of the horizontal axis periodicity fraction is greater than or equal to the total value of the vertical axis periodicity fraction, otherwise, determine that the image cutting direction is vertical cutting.
[0107] Optionally, the image segmentation region determination unit is specifically configured to:
[0108] Performing waveform extreme point detection processing on the transverse axis projection image to determine each waveform extreme point in the transverse axis projection image;
[0109] Isolating each waveform extreme point in the transverse axis projection image to obtain each isolated waveform image;
[0110] For each isolated waveform image, the original chip image is segmented according to the isolated waveform image to obtain an image segmentation area.
[0111] Optionally, the periodic total value determination unit is specifically configured to:
[0112] Determining a region segmentation direction of the image segmentation region, and determining a mutually perpendicular direction of the segmentation of the region segmentation direction;
[0113] Based on the mutually perpendicular directions of the segmentation, the pixel value statistical processing is performed on the segmented areas of the image to obtain a regional projection image;
[0114] Performing Fourier transform processing on the regional projection image to determine the regional main frequency corresponding to the image segmentation area;
[0115] The regional periodicity fraction corresponding to the image segmentation region is determined according to the main frequency amplitude of the regional main frequency.
[0116] Optionally, the first segmented image cropping module is specifically configured to:
[0117] Determine a median extreme point of all the waveform extreme points in the segmented projection image;
[0118] Constructing a period estimation interval based on the median extreme point and a predetermined preset waveform length;
[0119] The unit structure in the period estimation interval is subjected to periodicity detection processing to determine the estimated period length.
[0120] Optionally, the first segmented image cropping module is further specifically used for:
[0121] determining an image boundary of the first segmented image;
[0122] The first segmented image is cropped based on the estimated cycle length and the image boundary.
[0123] Optionally, the chip defect position determination module is specifically used to:
[0124] Determine a first evaluation index of the current second segmented image and a second evaluation index of the other second segmented images, wherein the first evaluation index at least includes a peak and trough value, a cross-correlation coefficient with other second segmented images, and a waveform similarity coefficient;
[0125] The first evaluation index is compared with each of the second evaluation indexes to determine the target second segmented image with defects.
[0126] The chip defect area determination device provided in the embodiment of the present invention can execute the chip defect area determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.
[0127] Embodiment 4
[0128] Fig.10 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.
[0129] like Fig.10 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.
[0130] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.
[0131] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the method for determining a chip defect area.
[0132] In some embodiments, the method chip defect area determination can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method chip defect area determination described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to perform the method chip defect area determination in any other appropriate manner (for example, by means of firmware).
[0133] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
[0134] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.
[0135] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0136] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).
[0137] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.
[0138] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.
[0139] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.
[0140] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A chip defect area determination method, characterized in that: include: Obtaining the original chip image corresponding to the target chip; Detecting the arrangement direction of the unit structure in the original chip image, determining the image cutting direction corresponding to the original chip image, and segmenting the original chip image according to the image cutting direction to obtain a first segmented image; For each of the first segmented images, based on directions perpendicular to each other of the image cutting directions, the first segmented images are subjected to pixel value statistical processing to obtain a segmented projection image, wherein the segmented projection image includes a plurality of waveform extreme value points, and the waveform extreme value points are peak extreme value points or trough extreme value points; Determine the estimated period length of the unit structure in the first segmented image according to the segmented projection image, and perform cropping processing on the first segmented image based on the estimated period length to obtain each second segmented image; For each of the second segmented images, compare the current second segmented image with other second segmented images to determine a target second segmented image with defects, and determine a defect position of the target chip based on the target second segmented image; The step of detecting the arrangement direction of the unit structure in the original chip image and determining the image cutting direction corresponding to the original chip image includes: Performing pixel value statistics processing on the chip pixel value image to obtain an initial projection image, wherein the initial projection image includes a horizontal axis projection image and a vertical axis projection image; Determining an image segmentation region of the original chip image according to the waveform extreme value points in the initial projection image; Determine the regional periodicity scores of the image segmentation regions in directions perpendicular to the segmentation directions, and determine the total image periodicity score value corresponding to the initial projection image according to the regional periodicity scores, wherein the total image periodicity score value includes the total periodicity score value of the horizontal axis and the total periodicity score value of the vertical axis; When the total value of the horizontal axis periodic fraction is greater than or equal to the total value of the vertical axis periodic fraction, the image cutting direction is determined to be horizontal cutting; otherwise, the image cutting direction is determined to be vertical cutting; The step of determining the image segmentation region corresponding to the original chip image according to the waveform extreme value points in the transverse projection image includes: Performing waveform extreme point detection processing on the transverse axis projection image to determine each waveform extreme point in the transverse axis projection image; Isolating each waveform extreme point in the transverse axis projection image to obtain each isolated waveform image; For each isolated waveform image, segmenting the original chip image according to the isolated waveform image to obtain an image segmentation area; The determining of the regional periodicity fractions of the image segmentation regions in directions perpendicular to the segmentation directions comprises: Determining a region segmentation direction of the image segmentation region, and determining a mutually perpendicular direction of the segmentation of the region segmentation direction; Based on the mutually perpendicular directions of the segmentation, the pixel value statistical processing is performed on the segmented areas of the image to obtain a regional projection image; Performing Fourier transform processing on the regional projection image to determine the regional main frequency corresponding to the image segmentation area; The regional periodicity fraction corresponding to the image segmentation region is determined according to the main frequency amplitude of the regional main frequency.
2. The method according to claim 1, characterized in that The step of determining the estimated period length of the unit structure in the first segmented image according to the segmented projection image comprises: Determine a median extreme point of all the waveform extreme points in the segmented projection image; Constructing a period estimation interval based on the median extreme point and a predetermined preset waveform length; The cycle estimation interval is subjected to periodic detection processing to determine the estimated cycle length.
3. The method according to claim 2, characterized in that The step of performing a cropping process on the first segmented image based on the estimated cycle length includes: determining an image boundary of the first segmented image; The first segmented image is cropped based on the estimated cycle length and the image boundary.
4. The method according to claim 1, characterized in that: The step of comparing the current second segmented image with other second segmented images to determine a target second segmented image with defects includes: Determine a first evaluation index of the current second segmented image and a second evaluation index of the other second segmented images, wherein the first evaluation index at least includes a peak and trough value, a cross-correlation coefficient with other second segmented images, and a waveform similarity coefficient; The first evaluation index is compared with each of the second evaluation indexes to determine the target second segmented image with defects.
5. A chip defect area determination device, characterized in that: include: An original chip image acquisition module is used to acquire the original chip image corresponding to the target chip; An original chip image segmentation module is used to detect the arrangement direction of the unit structure in the original chip image, determine the image cutting direction corresponding to the original chip image, and segment the original chip image according to the image cutting direction to obtain a first segmented image; a segmented projection image determination module, configured to perform pixel value statistical processing on each of the first segmented images based on directions perpendicular to each other of the image cutting directions, to obtain a segmented projection image, wherein the segmented projection image includes a waveform extreme value point, and the waveform extreme value point is a peak extreme value point or a trough extreme value point; A first segmented image cropping module, configured to determine an estimated period length of a unit structure in the first segmented image according to the segmented projection image, and to crop the first segmented image based on the estimated period length to obtain respective second segmented images; a chip defect position determination module, configured to compare the current second segmented image with other second segmented images for each of the second segmented images, determine a target second segmented image with defects, and determine the defect position of the target chip based on the target second segmented image; The original chip image segmentation module comprises: An initial projection image determination unit, used for performing pixel value statistical processing on the original chip image to obtain an initial projection image, wherein the initial projection image includes a horizontal axis projection image and a vertical axis projection image; An image segmentation region determination unit, used to determine the image segmentation region corresponding to the original chip image according to the waveform extreme value points in the initial projection image; a periodicity total value determination unit, configured to determine a regional periodicity score of the image segmentation region in a direction perpendicular to the segmentation direction, and determine a total image periodicity score value corresponding to the initial projection image according to the regional periodicity score, wherein the total image periodicity score value includes a total horizontal axis periodicity score value and a total vertical axis periodicity score value; An image cutting direction determining unit, configured to determine that the image cutting direction is a horizontal cutting if the total value of the horizontal axis periodicity fraction is greater than or equal to the total value of the vertical axis periodicity fraction, and otherwise, determine that the image cutting direction is a vertical cutting; The image segmentation region determination unit is specifically used for: Performing waveform extreme point detection processing on the transverse axis projection image to determine each waveform extreme point in the transverse axis projection image; Isolating each waveform extreme point in the transverse axis projection image to obtain each isolated waveform image; For each isolated waveform image, segmenting the original chip image according to the isolated waveform image to obtain an image segmentation area; The periodic total value determination unit is specifically used for: Determining a region segmentation direction of the image segmentation region, and determining a mutually perpendicular direction of the segmentation of the region segmentation direction; Based on the mutually perpendicular directions of the segmentation, the pixel value statistical processing is performed on the segmented areas of the image to obtain a regional projection image; Performing Fourier transform processing on the regional projection image to determine the regional main frequency corresponding to the image segmentation area; The regional periodicity fraction corresponding to the image segmentation region is determined according to the main frequency amplitude of the regional main frequency.
6. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the chip defect area determination method according to any one of claims 1 to 4.
7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the chip defect area determination method according to any one of claims 1 to 4 when executed.
