Method and device for identifying unsealing of chip

Abstract

The invention is suitable for the technical field of image recognition, and provides a method and device for recognizing unsealing of a chip, and the method comprises the steps: collecting a current real-time image, rotating the current real-time image to a preset direction, and obtaining a current chip image; obtaining a region of interest in the standard template image; wherein the region of interest comprises a bare chip region, a wire bonding region and / or a pin region; extracting a to-be-identified region corresponding to the region of interest in the current chip image; and controlling a chip unsealing process according to the region similarity between the region of interest and the region to be identified corresponding to the region of interest. According to the scheme, high efficiency, accuracy and automation in the chip unsealing process are achieved, and the unsealing efficiency and the recognition accuracy are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of image recognition, and particularly relates to a method and device for recognizing chip opening. BACKGROUND

[0002] In the process of electronic device manufacturing and maintenance, chip opening is a key step. Chip opening is usually performed for fault analysis, quality detection or other maintenance operations. However, identification and positioning during chip opening has always been a challenge. Traditional chip opening methods mainly rely on manual operation and visual inspection, which is not only time-consuming and labor-intensive, but also difficult to guarantee accuracy and efficiency.

[0003] With the development of technology, image processing technology and machine learning algorithms have been widely applied in various fields, including the automation of chip opening. Through image recognition technology, the identification accuracy and efficiency of the chip opening process can be significantly improved, reducing human error and operation time. However, existing automated chip opening identification methods still have some shortcomings. Existing methods often cannot accurately position the region of interest, affecting the accuracy of subsequent identification. SUMMARY

[0004] Therefore, the embodiments of the present application provide a method and device for recognizing chip opening to solve the technical problem that existing methods often cannot accurately position the region of interest.

[0005] A first aspect of the embodiments of the present application provides a method for recognizing chip opening, which comprises: collecting a current real-time image and rotating the current real-time image to a preset direction to obtain a current chip image; wherein the preset direction refers to the direction corresponding to the standard chip region in the standard template image; obtaining a region of interest in a standard template image; wherein the region of interest includes a chip die region, a wire bonding region and / or a pin region; extracting a to-be-identified region corresponding to the region of interest in the current chip image; controlling the chip opening process according to the region similarity between the region of interest and the to-be-identified region corresponding to the region of interest.

[0006] Further, the step of collecting a current real-time image and rotating the current real-time image to a preset direction to obtain a current chip image comprises: collecting an unopened chip image and extracting an unopened chip region in the unopened chip image; extracting a current identification pattern in the unopened chip region; wherein the current identification pattern refers to a product identification, logo or specific pattern printed on the chip packaging region; extracting a current direction of the current identification pattern; During the chip opening process, a current real-time image is collected, and the current real-time image is rotated to a preset direction based on the current direction to obtain a current chip image.

[0007] Further, the step of extracting the current direction of the current identification pattern comprises: extracting a standard identification pattern in a standard template image; constructing a first coordinate system with the center of the standard packaging area as the origin and the length and width directions as the axes; Based on the first coordinate system, a plurality of first sampling points on the edge of the standard identification pattern are extracted at a fixed horizontal coordinate interval within a preset angle range; wherein the preset angle range includes 0 degrees to 90 degrees, 90 degrees to 180 degrees, 180 degrees to 270 degrees, or 270 degrees to 360 degrees; constructing a second coordinate system with the center of the current identification pattern as the origin and the length and width directions as the axes; Based on the second coordinate system, a plurality of second sampling points on the edge of the current identification pattern are extracted at a fixed horizontal coordinate interval within a plurality of current angle ranges; wherein the plurality of current angle ranges include 0 degrees to 90 degrees, 90 degrees to 180 degrees, 180 degrees to 270 degrees, and 270 degrees to 360 degrees; According to the plurality of first sampling points and the plurality of second sampling points, the current direction of the current identification pattern is calculated.

[0008] Further, the step of calculating the current direction of the current identification pattern according to the plurality of first sampling points and the plurality of second sampling points comprises: calculating a plurality of first distances between the plurality of first sampling points and the origin, and constructing the plurality of first distances as a first distance vector; calculating a plurality of second distances between the plurality of second sampling points and the origin, and constructing the plurality of second distances as a second distance vector; respectively calculating the vector similarity between the second distance vector corresponding to each of the plurality of current angle ranges and the first distance vector; calculating the direction included angle between the current angle range corresponding to the maximum vector similarity and the preset angle range; The direction included angle is taken as the current direction.

[0009] Further, the step of controlling the chip opening process according to the region similarity between the region of interest and the to-be-identified region corresponding to the region of interest comprises: calculating the region similarity between the region of interest and the to-be-identified region corresponding to the region of interest; When the region similarity is lower than the preset similarity, continue to execute the chip opening process, and collect subsequent real-time images; When the region similarity is lower than the preset similarity, continue to execute the chip opening process, and collect subsequent real-time images;

[0010] Further, the step of calculating the region similarity between the region of interest and the to-be-identified region corresponding to the region of interest comprises: extracting a first pixel value and a second pixel value corresponding to the same pixel point position in the region of interest and the to-be-identified region corresponding to the region of interest; calculating an absolute value difference between the first pixel value and the second pixel value; if the absolute value difference is less than a first value, the pixel point position is taken as a pixel position of interest; if the absolute value difference is not less than the first value, the pixel point position is taken as a non-pixel position of interest; extracting a neighboring pixel position of interest corresponding to each pixel position of interest, to obtain a current image region composed of multiple neighboring pixel positions of interest or a single pixel position of interest; counting the number of pixel positions in the current image region; taking the current image region with a number of pixel positions greater than a second number as a target image region; calculating the region similarity according to the number of pixel positions corresponding to the target image region and the absolute value difference corresponding to each pixel point position.

[0011] Further, the step of calculating the region similarity according to the number of pixel positions corresponding to the target image region and the absolute value difference corresponding to each pixel point position comprises: obtaining a maximum absolute value difference; calculating a quality weight corresponding to each pixel point position according to the maximum absolute value difference and the absolute value difference corresponding to each pixel point position; wherein the quality weight is , di i , di max represents the maximum absolute value difference, represents a decay coefficient; calculating a region contribution degree according to the number of pixel positions corresponding to the target image region and the quality weight corresponding to each pixel point position; wherein the region contribution degree is , nj j , nj i represents the quality weight corresponding to the i-th pixel point position; taking the average value of the region contribution degrees corresponding to multiple target image regions as the region similarity.

[0012] The second aspect of the embodiment of the present application provides a device for identifying chip opening, comprising: An acquisition unit is configured to acquire a current real-time image and rotate the current real-time image to a preset direction to obtain a current chip image; wherein the preset direction refers to a direction corresponding to a standard chip region in a standard template image; An acquisition unit is configured to acquire a region of interest in a standard template image; wherein the region of interest includes a chip die region, a wire bonding region, and / or a packaging pin region; An extraction unit is configured to extract a to-be-identified region corresponding to the region of interest in the current chip image; A control unit is configured to control a chip opening process according to a region similarity between the region of interest and the to-be-identified region corresponding to the region of interest.

[0013] The third aspect of the embodiment of the present application provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in the method for identifying chip opening according to the first aspect when executing the computer program.

[0014] The fourth aspect of the embodiment of the present application provides a computer readable storage medium, which stores a computer program, wherein the computer program is executed by a processor to implement the steps in the method for identifying chip opening according to the first aspect.

[0015] Compared with the prior art, the embodiment of the present application has the beneficial effects that: by rotating the current real-time image to the preset direction, it is ensured that the image direction is consistent each time. Since the chip image may have different features in different directions, the unified image direction helps to extract and match the image features, thereby improving the accuracy and consistency of the identification. Using the standard template image as a reference ensures that there is a reliable benchmark during the identification process, thereby reducing errors caused by different image directions and angles. By acquiring the region of interest (such as the chip die region, the wire bonding region, and the pin region) in the standard template image and extracting the corresponding to-be-identified region in the current chip image, the key parts of the chip can be accurately positioned. This process avoids the identification errors caused by inaccurate extraction regions in the traditional method. The region similarity between the region of interest and the corresponding to-be-identified region is used for judgment, which further improves the accuracy of the identification. The similarity calculation can effectively distinguish the subtle differences, thereby ensuring the reliability of the identification result. In summary, the present application realizes efficient, accurate, and automated chip opening process, significantly improving the opening efficiency and identification accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A schematic flowchart of a method for identifying chip unpacking provided by the present invention is shown; Figure 2 This diagram illustrates a device for identifying chip unpacking according to an embodiment of the present invention; Figure 3 A schematic diagram of a terminal device provided in an embodiment of the present invention is shown. Detailed Implementation

[0018] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0019] This invention provides a method and apparatus for identifying chip unpacking, in order to solve the technical problem that existing methods often cannot accurately locate the region of interest.

[0020] First, this invention provides a method for identifying chip unpacking. Please refer to [link / reference]. Figure 1 , Figure 1 A schematic flowchart of a method for identifying chip unpacking provided by the present invention is shown. Figure 1 As shown, the method for unlocking the identification chip may include the following steps: Step 101: Acquire the current real-time image and rotate the current real-time image to a preset direction to obtain the current chip image; wherein, the preset direction refers to the direction corresponding to the standard chip area in the standard template image; Since there may be inconsistencies in chip orientation during the chip unpacking process, it is necessary to rotate and transform the acquired real-time image to make its orientation completely consistent with the chip orientation in a pre-set "standard template image".

[0021] The standard template image is a reference image of a known high-quality chip that is in a normal, unopened state. The preset orientation is the orientation of the chip region (chip body) within this standard template image. This ensures spatial comparability between the current chip image and the standard template image. Subsequent feature extraction and similarity comparison rely on this orientational alignment.

[0022] Specifically, step 101 includes steps 1011 to 1014: Step 1011: Acquire an image of the unopened chip and extract the unopened chip region from the image; Before the chip unpacking process begins (i.e., when the chip is completely unpacked), an image of the specific chip is first acquired. Then, edge detection is used to identify and segment the area containing the chip itself (i.e., the "unpacked chip area") from this entire image. This area includes the package itself, excluding background, fixtures, and other irrelevant parts. Obtaining a clean image area containing the target chip (especially its package) provides the basis for subsequent steps.

[0023] Step 1012: Extract the current identification graphic from the unopened chip area; wherein, the current identification graphic refers to the product identification, mark or specific pattern printed on the chip packaging area; Precisely locate and isolate the current identification graphic on the package. The current identification graphic refers to the product identification, mark, or specific pattern printed on the chip package area.

[0024] The current identifier graphics have clear, stable, and easily detectable geometric features (such as edges, corners, and specific markings). Segmentation based on color / texture is an existing technique and will not be described further here.

[0025] Step 1013: Extract the current orientation of the current identifier graphic; Analyze the current logo obtained in the previous step, calculate and determine its orientation angle (i.e., "current direction") in the image coordinate system.

[0026] Specifically, step 1013 includes steps A1 to A6: Step A1: Extract the standard logo graphic from the standard template image; The extraction logic for standard identification graphics is the same as that for current identification graphics, and will not be repeated here. Standard identification graphics refer to product identification, logos, or specific patterns printed on the chip packaging area.

[0027] Step A2: Construct a first coordinate system with the center of the standard package area as the origin and the length and width directions as axes; A reference coordinate system is established which is closely related to the geometry of the standard package area. Subsequent point sampling will be performed in this coordinate system, ensuring that the sampling point positions are stable relative to the center of the package area.

[0028] Step A3: Based on the first coordinate system, a plurality of first sampling points on the edge of the standard identification pattern are extracted at a fixed horizontal coordinate interval within a preset angle range; wherein the preset angle range includes 0 degrees to 90 degrees, 90 degrees to 180 degrees, 180 degrees to 270 degrees, or 270 degrees to 360 degrees. The preset angle range is one of the four 90-degree quadrants (0-90°, 90-180°, 180-270°, 270-360°). This means that the system only selects one 90-degree quadrant (for example, the lower left quadrant) for sampling. Within the selected 90-degree quadrant, move along the X-axis direction of the first coordinate system at a fixed pixel interval (or physical interval). For each moved horizontal coordinate position (x_i), find the point intersecting the edge of the standard identification pattern along the direction from the center to the outside on the vertical line corresponding to the horizontal coordinate, as the sampling point P_std_i. This will obtain a set of sampling points {P_std_1, P_std_2,..., P_std_N} which are all located on the boundary of the selected 90-degree quadrant.

[0029] A set of stable, position-known edge reference points are obtained on a specific local area (a quadrant) of the standard identification pattern. Selecting a quadrant can reduce the amount of calculation and may avoid interference caused by symmetry.

[0030] Step A4: A second coordinate system is constructed with the center of the current identification pattern as the origin and the length and width directions as the axes. A local coordinate system is established for the current identification pattern, with its axes aligned with the estimated main and secondary axes, in preparation for sampling on its edge. This coordinate system is independent of its global rotation angle.

[0031] Step A5: Based on the second coordinate system, a plurality of second sampling points on the edge of the current identification pattern are extracted at a fixed horizontal coordinate interval within a plurality of current angle ranges; wherein the plurality of current angle ranges include 0 degrees to 90 degrees, 90 degrees to 180 degrees, 180 degrees to 270 degrees, and 270 degrees to 360 degrees. The plurality of current angle ranges include all four 90-degree quadrants (0-90°, 90-180°, 180-270°, 270-360°). This means that the system will sample on the edge of each 90-degree quadrant of the current identification pattern separately.

[0032] In each quadrant, move along the X-axis direction of the second coordinate system at the same fixed interval as step A3.

[0033] For each quadrant and each moved-to horizontal coordinate position (x_j), find the point intersecting the current identification pattern edge on the vertical line corresponding to the horizontal coordinate as a sampling point. Finally, four sets of sampling points covering the entire edge of the package region are obtained: {P_cur_quad1}, {P_cur_quad2}, {P_cur_quad3}, {P_cur_quad4}.

[0034] Obtain a set of uniformly distributed sampling points on the entire edge of the current identification pattern. The sampling strategy (fixed horizontal coordinate interval) is consistent with that of the standard template.

[0035] Step A6: calculating the current direction of the current identification pattern according to the plurality of first sampling points and the plurality of second sampling points.

[0036] By comparing the spatial relationship between the local reference point set (one quadrant) of the standard package region and the global point set (entire edge) of the current identification pattern, the rotation angle difference (i.e. the current direction) of the current identification pattern relative to the standard package region is accurately calculated.

[0037] In the embodiments corresponding to steps A1 to A6, the local reference points of one quadrant of the standard template ({P_std}) are used to match the global point set ({P_cur}) of the complete edge of the current identification pattern. The matching process needs to determine the best matching quadrant and the best rotation angle at the same time. By utilizing the rich geometric information of the identification pattern (multiple sampling points) and minimizing the error by optimizing the matching, an accurate rotation angle θ at the sub-pixel level can be calculated, which is significantly better than the method of relying only on the main axis or a few feature points. For nearly rectangular or square packages, there may be 90°, 180°, and 270° symmetry confusion. By forcing the matching of a specific quadrant (the standard template only takes one quadrant) and searching for the best rotation and quadrant correspondence, these symmetry directions can be effectively distinguished, avoiding 180° errors. If the current chip is rotated by 90°, the actual edge quadrant shape corresponding to the standard template sampling quadrant position in {P_cur} will be completely different, and the matching distance will be large, while the shape matching degree of the adjacent quadrant after rotation by 90° will be higher, so that the correct 90° deviation can be calculated. By densely sampling the edge points of the identification pattern (especially for global sampling of the current package) and performing point set matching optimization, an angle resolution at the sub-pixel level is achieved. The standard template only needs to store the point set of one quadrant, reducing the computational burden, while the global sampling of the current package ensures the accuracy and robustness of the matching.

[0038] Specifically, step A6 specifically includes steps A61 to A65: Step A61: Calculate a plurality of first distances between a plurality of first sampling points and an origin point, and construct the plurality of first distances as a first distance vector; For each first sampling point P_std_i extracted in the standard template image (these points are all located in the same selected preset angle range / quadrant, for example, the 0-90° quadrant), the Euclidean distance d_std_i = ||P_std_i - O_std|| of the point to its first coordinate system origin point O_std (i.e., the center of the standard packaging region) is calculated.

[0039] According to the fixed order of the sampling points in the preset angle range (quadrant) (for example, arranged in order of fixed horizontal coordinate interval), all the distance values d_std_1, d_std_2,..., d_std_N calculated are combined into an ordered list or array. This ordered set is the first distance vector V_std.

[0040] The geometric shape information (distance distribution of sampling points to the center) of the standard template in a specific quadrant is abstracted into a one-dimensional numerical sequence (vector). This vector describes the "radius" variation characteristics of the quadrant edge contour relative to the center.

[0041] Step A62: Calculate a plurality of second distances between a plurality of second sampling points and an origin point, and construct the plurality of second distances as a second distance vector; For each second sampling point P_cur_j extracted in the current identification pattern image (these points are distributed in all four current angle ranges / quadrants), the Euclidean distance d_cur_j = ||P_cur_j - O_cur|| of the point to its second coordinate system origin point O_cur (i.e., the center of the current identification pattern) is calculated.

[0042] The global sampling points of the current identification pattern (distributed in four quadrants) are grouped by quadrant. For each quadrant q (q = 1, 2, 3, 4, corresponding to 0-90°, 90-180°, 180-270°, 270-360°, respectively), all the sampling points {P_cur_q} belonging to the quadrant q are taken out. According to the fixed order of these points in the quadrant q (the same rule as the standard template sampling, such as arranged in order of fixed horizontal coordinate interval), the distances of these points to O_cur are calculated {d_cur_q}. These distance values are combined in order to form an ordered list or array, forming the second distance vector V_cur_q of the quadrant q.

[0043] Finally, four second distance vectors are obtained: V_cur_quad1, V_cur_quad2, V_cur_quad3, V_cur_quad4, each of which represents the contour distance feature of a complete quadrant.

[0044] The geometric shape information of each complete quadrant of the current identification pattern is also abstracted into a respective one-dimensional distance vector. This is done in preparation for subsequent quadrant matching.

[0045] Step A63: Calculate the vector similarity between the second distance vector corresponding to each of the plurality of current angle ranges and the first distance vector, respectively; For each current angle range (i.e., each quadrant q), compare its corresponding second distance vector V_cur_q with the first distance vector V_std of the standard template.

[0046] Calculate the vector similarity Sim_q between them. The vector similarity is calculated by cosine similarity or Pearson correlation coefficient.

[0047] Quantitatively evaluate the matching degree of the edge contour feature of each complete quadrant of the current identification pattern (denoted by V_cur_q) and the edge contour feature of the specific reference quadrant of the standard template (denoted by V_std). The higher the similarity Sim_q, the more similar the shape feature of the current q-th quadrant is to the standard reference quadrant.

[0048] Step A64: Calculate the directional angle between the current angle range corresponding to the maximum vector similarity and the preset angle range; Find the maximum value Sim_max among the four quadrant similarity values (Sim_quad1, Sim_quad2, Sim_quad3, Sim_quad4) calculated in step A63. Determine the current angle range (quadrant) q_max corresponding to this maximum value (for example, if Sim_quad3 is the maximum, then q_max = Quad3, corresponding to the angle range 180-270°).

[0049] The sampling points of the standard template only come from one preset angle range (quadrant) q_std (for example, the preset is Quad1, 0-90°). Compare the angle positions of q_max (the current quadrant with the highest matching degree) and q_std (the reference quadrant of the standard template). The directional angle θ is the angle difference between the two quadrants. The specific calculation method depends on the definition of the quadrant number / order: If the quadrants are numbered in counter-clockwise order (Quad1: 0-90°, Quad2: 90-180°, Quad3: 180-270°, Quad4: 270-360°), then θ= (q_max_index - q_std_index) * 90°.

[0050] For example: q_std = Quad1 (index=1), q_max = Quad3 (index=3), then θ= (3 - 1) *90° = 180°.

[0051] If the result is negative or greater than 270°, it may need to be adjusted to the range of [-180°, 180°] or [0°, 360°) (for example, -90° is equivalent to 270°), and the minimum rotation angle is taken.

[0052] By finding the current quadrant that is most similar to the standard reference quadrant, the discrete rotation angle offset (multiple of 90°) of the current identification pattern relative to the standard template is directly determined.

[0053] Step A65: Take the direction angle as the current direction.

[0054] In the embodiments corresponding to steps A61 to A65, based on fast distance calculation and one-dimensional vector similarity comparison, high real-time requirements are met. The translation in the image plane is completely invariant, the 90° multiple rotation is perfectly identified, and there is a certain tolerance for local edge noise and global illumination / slight scale changes. This method is particularly suitable as a fast direction correction step for processing rectangular / square chips in the common scenario of standard discrete angles (0°, 90°, 180°, 270°) placement in a chip opening vision guidance or detection system, providing a basis for subsequent accurate ROI extraction and state analysis. A highly efficient and simple algorithm is provided for determining the 90° multiple rotation angle (current direction) of the chip packaging area relative to the standard template. The core is to convert the shape information of the packaging edge into a one-dimensional distance vector, and match the best rotation posture through quadrant similarity comparison.

[0055] Step 1014: During the chip opening process, a current real-time image is captured, and the current real-time image is rotated to a preset direction based on the current direction, to obtain a current chip image.

[0056] A rotation transform is applied to the captured real-time image. The angle of rotation is: reverse-rotating the image by the angle value of the "current direction" (or equivalently, forward-rotating by an angle that zeros the direction). This compensates for the random orientation of the chip at initial placement, aligning it with the preset direction. Using the direction reference calculated from the same chip before opening, the direction problem of the chip that may have moved slightly during opening or was initially placed with a slight deviation is corrected in real time, ensuring that the subsequent ROI extraction and similarity comparison with the standard template image are in the same direction.

[0057] Only when the current real-time image is accurately rotated to the same preset direction as the standard template image, can the ROI coordinates defined from the standard template be accurately mapped to the corresponding physical location on the current image. Direction alignment is the basis for spatial coordinate matching.

[0058] When the images of the two regions are strictly aligned in direction and position, comparing their visual similarity (whether at the pixel level, feature level, or structure level) has reliable meaning. Direction inconsistency can introduce huge differences unrelated to the opening state, making the similarity result unreliable and leading to false positives / negatives.

[0059] In the embodiments corresponding to steps 1011 to 1014, a direction reference based on the individual package features of each chip is extracted for each chip, and this individualized reference is used to perform high-precision direction correction on the real-time image during opening. The direction deviation caused by the initial placement of the chip and the slight rotation during opening is effectively eliminated, ensuring that the subsequent key region extraction and state comparison are performed in a strictly aligned spatial coordinate system, thereby greatly improving the accuracy, robustness, and automation level of the chip opening state recognition method.

[0060] Step 102: Obtain a region of interest in the standard template image; wherein the region of interest includes a chip die area, a wire bonding area, and / or a pin area; In the standard template image, one or more specific sub-regions (Region of Interest, ROI) are pre-calibrated. The ROI is a key structural part in the chip opening process, including but not limited to the chip die area, the wire bonding area, and / or the pin area. The chip die area is the silicon wafer part of the chip core, which should be exposed after opening. The wire bonding area is the small metal wire area connecting the solder points on the chip die and the lead frame, which should be clearly visible and undamaged after opening. The pin area is the metal pin part of the chip that connects to the external circuit.

[0061] Step 103: Extract the to-be-recognized region corresponding to the region of interest in the current chip image; Locate and extract the region corresponding to the template ROI on the current chip image after alignment of the direction. According to the mapping relationship, cut out a sub-image region from the current chip image which strictly corresponds to the standard template ROI in position (i.e. the same coordinate position), as the "to-be-recognized region" to be detected.

[0062] Step 104: Control the chip opening process according to the region similarity between the region of interest and the to-be-recognized region corresponding to the region of interest.

[0063] For each "region of interest" defined in step 102 and extracted in step 103 (in the template map) and its "corresponding to-be-recognized region" (in the current map), calculate the visual similarity degree (similarity score) between them. Based on the calculated similarity (possibly the similarity of one or more regions), decide the next action of the opening device.

[0064] Specifically, step 104 specifically includes steps 1041 to 1043: Step 1041: Calculate the region similarity between the region of interest and the to-be-recognized region corresponding to the region of interest; Quantify the difference between the current visual state of each key region (die, bonding wire, pin) after opening and the standard template (normal state). The lower the region similarity, the greater the difference between the state of the region and the normal state.

[0065] Specifically, step 1041 specifically includes steps B1 to B8: Step B1: Extract the first pixel value and the second pixel value corresponding to the same pixel point position in the region of interest and the to-be-recognized region corresponding to the region of interest; The first pixel value is the pixel value of the point in the standard template image. The second pixel value is the pixel value of the corresponding point in the current to-be-recognized region.

[0066] Step B2: Calculate the absolute value difference between the first pixel value and the second pixel value; Subtract the first pixel value and the second pixel value corresponding to the same pixel point position, and take the absolute value to get the absolute value difference. By calculating the absolute value difference, the difference between each pixel point can be accurately measured, providing more accurate similarity analysis.

[0067] Step B3: If the absolute value difference is less than a first value, the pixel point position is taken as the region of interest pixel position; If the difference is less than a certain threshold (first value), it is considered that the pixel point position is similar in the two images, and is taken as the region of interest pixel position. By setting the threshold, similar pixel points can be effectively filtered out, thereby improving the accuracy of the similarity calculation. The first value is 10.

[0068] Step B4: If the absolute value difference is not less than the first value, the pixel position is marked as a non-interesting pixel position. If the difference is greater than or equal to the first value, it is considered that the pixel position is not similar in the two images, and is marked as a non-interesting pixel position.

[0069] Step B5: Extract the adjacent interesting pixel positions corresponding to each interesting pixel position to obtain a current image region composed of multiple adjacent interesting pixel positions or a single interesting pixel position. After determining the interesting pixel position, the adjacent interesting pixel positions around it are extracted to form a larger region. By aggregating adjacent interesting pixel positions, the similarity of the image region can be more comprehensively reflected.

[0070] Step B6: Count the number of pixel positions in the current image region. Step B7: The current image region with a pixel position number greater than the second number is regarded as a target image region. If the number of interesting pixels in the current image region exceeds a certain threshold (the second number), the region is determined as a target image region. By setting the threshold, regions containing a large number of similar pixels can be screened out, improving the effectiveness of similarity calculation.

[0071] If the second number is 30, and the current image region has 50 interesting pixel positions, it is marked as a target image region.

[0072] Step B8: Calculate the region similarity according to the number of pixel positions corresponding to the target image region and the absolute value difference corresponding to each pixel position.

[0073] Comprehensive use of pixel number and difference value provides more comprehensive and accurate region similarity measurement, thereby improving the accuracy of image matching and recognition.

[0074] In the embodiments corresponding to steps B1 to B8, by setting specific values (such as the first value being 10 and the second number being 30), the steps and standards of similarity calculation can be accurately controlled to ensure the efficiency and accuracy of the identification process. These logical steps and values not only improve the fineness of similarity calculation, but also effectively filter and aggregate similar image regions, thereby enhancing the reliability and security of the entire chip opening process.

[0075] Specifically, step B8 specifically includes steps B81 to B84: Step B81: Obtain the maximum absolute value difference. Step B82: Calculate the quality weight corresponding to each pixel position according to the maximum absolute value difference and the absolute value difference corresponding to each pixel position; wherein the quality weight = w d i represents the absolute value difference corresponding to the i-th pixel position, d max represents the maximum absolute value difference, represents the decay coefficient; More in line with human visual sensitivity to differences (high tolerance for small differences, strong punishment for large differences) by Controlling sensitivity (e.g., setting Lower rigor.

[0076] Step B83: Calculate the region contribution degree according to the number of pixel positions corresponding to the target image region and the quality weight corresponding to each pixel position; wherein the region contribution degree = w , n j represents the number of pixel positions corresponding to the j-th target image region, w i represents the quality weight corresponding to the i-th pixel position; For each target image region, calculate the weighted effective pixel number, and the average quality weight in the parentheses. The region contribution degree not only considers the average similarity of the region, but also considers the scale of the region. In this way, the contribution between small-scale but highly similar regions and large-scale but lowly similar regions can be balanced.

[0077] Step B84: Take the average of the region contribution degrees corresponding to the plurality of target image regions as the region similarity.

[0078] By calculating the average of the region contribution degrees of all target image regions, a comprehensive region similarity score can be obtained. This score reflects the overall similarity between the region of interest and the region to be identified.

[0079] Since the region contribution degree takes into account both the similarity of the pixel points in the region and the scale of the region, the final region similarity can more comprehensively reflect the similarity relationship between the two regions, reducing the deviation caused by a single factor.

[0080] In the embodiments corresponding to steps B81 to B84, by standardizing pixel differences, exponentially decaying weights, and comprehensively considering region scale and similarity, the similarity between the region of interest and the region to be identified can be effectively evaluated. It not only considers the detailed differences at the pixel level, but also ensures the rationality and robustness of large-scale statistics through the calculation of region contribution degree, thereby providing an accurate and comprehensive region similarity calculation method.

[0081] Step 1042: When the similarity of the region is lower than the preset similarity, continue to execute the chip unpacking process and acquire subsequent real-time images; Send instructions to the opening equipment (such as a laser opening machine or a chemical opening tank controller) to maintain or continue the opening operation (e.g., maintain laser power, continue injecting etching solution, move to the next opening point, etc.). After the opening process continues for a period of time (or at the next preset detection point), acquire a new real-time image and execute steps 101 to 104.

[0082] When the area similarity is lower than the preset similarity, it means that the sealant has not been completely removed, the bare die has not been fully exposed, or there are still obstructions in the bonding wire area. Therefore, the system decides to continue the unsealing operation to improve the condition of these areas (make them more similar to the standard template) in subsequent steps.

[0083] Acquiring subsequent images is for the purpose of the next detection-decision loop, forming a closed-loop control. The system will continue to monitor until all areas meet the requirements.

[0084] Step 1043: When the similarity of the regions is not lower than the preset similarity, stop the chip unpacking process.

[0085] When the similarity of all regions is not lower than the preset similarity, it means that the status of all monitored key regions has reached or exceeded the expected normal state (the similarity of all regions is not lower than the threshold). Stop production as soon as the quality requirements are met (all regions meet the standards) to ensure product quality (avoid overprocessing damage) and optimize the production cycle (do not waste time on completed steps).

[0086] In the embodiments corresponding to steps 1041 to 1043, instead of relying on preset fixed time, fixed energy, or fixed steps, the decision to continue unpacking is dynamically made based on the current actual visual state of each critical area (quantified by similarity measurement). This automatically adapts to minor differences between different chips, changes in the state of the unpacking tool (such as laser power attenuation, etchant concentration changes), and environmental fluctuations. Unpacking only stops when all critical areas meet the requirements, ensuring complete unpacking and quality compliance. If even one area fails to meet the standard, unpacking continues to prevent under-unpacking (outflow of defective products); unpacking stops immediately when all areas meet the standard, preventing over-unpacking (damage to the chip, waste of resources). The process of "continuing to execute...and acquiring subsequent real-time images" constitutes a closed-loop control loop, continuously monitoring and adjusting until all conditions are met. By precisely controlling the unpacking endpoint, the integrity and cleanliness of critical structures (bare die, bonding wires, pins) after unpacking are ensured. Unpacking stops immediately upon meeting quality requirements, avoiding unnecessary additional unpacking time / energy / material consumption.

[0087] In the embodiments corresponding to steps 101 to 104, by rotating the current real-time image to the preset direction, the image direction is consistent for each processing. Since the chip image may have different features in different directions, uniform image direction helps to extract and match image features, improving the accuracy and consistency of recognition. Using a standard template image as a reference ensures that there is always a reliable benchmark during the recognition process, thereby reducing errors due to different image directions and angles. By obtaining the region of interest (such as the chip die area, wire bonding area, and pin area) in the standard template image and extracting the corresponding to-be-recognized region in the current chip image, the key parts of the chip can be accurately positioned. This process avoids recognition errors caused by inaccurate extraction area in traditional methods. According to the region similarity between the region of interest and the corresponding to-be-recognized region, the accuracy of recognition is further improved. Similarity calculation can effectively distinguish subtle differences, thereby ensuring the reliability of the recognition result. In summary, the present application realizes efficient, accurate and automated chip opening process, significantly improving the opening efficiency and recognition accuracy.

[0088] As Figure 2 The present application provides a device for identifying chip opening, please see Figure 2 , Figure 2 A device for identifying chip opening provided by the present application is shown in the schematic diagram, as Figure 2 shown, a device for identifying chip opening includes: The acquisition unit 21 is used for acquiring a current real-time image, and rotating the current real-time image to a preset direction to obtain a current chip image; wherein the preset direction refers to the direction corresponding to the standard chip region in the standard template image; The acquisition unit 22 is used for acquiring a region of interest in the standard template image; wherein the region of interest includes a chip die area, a wire bonding area, and / or a packaging pin area; The extraction unit 23 is used for extracting a to-be-recognized region corresponding to the region of interest in the current chip image; The control unit 24 is used for controlling the chip opening process according to the region similarity between the region of interest and the to-be-recognized region corresponding to the region of interest.

[0089] This invention provides a device for identifying chip unpacking. By rotating the current real-time image to a preset orientation, it ensures consistent image orientation for each processing operation. Since chip images may exhibit different features in different orientations, a unified image orientation facilitates feature extraction and matching, improving the accuracy and consistency of identification. Using a standard template image as a reference ensures a reliable benchmark throughout the identification process, reducing errors caused by variations in image orientation and angle. By acquiring regions of interest (such as the die area, wire bonding area, and pin area) from the standard template image and extracting the corresponding region to be identified in the current chip image, the key components of the chip can be accurately located. This process avoids identification errors caused by inaccurate region extraction in traditional methods. Judgment is based on the similarity between the region of interest and the corresponding region to be identified, further improving identification accuracy. Similarity calculation effectively distinguishes subtle differences, ensuring the reliability of the identification results. In summary, this invention achieves high efficiency, accuracy, and automation in the chip unpacking process, significantly improving unpacking efficiency and identification accuracy.

[0090] Figure 3 This is a schematic diagram of a terminal device provided in an embodiment of the present invention. Figure 3 As shown, a terminal device 3 in this embodiment includes: a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30, such as a program for identifying chip unpacking. When the processor 30 executes the computer program 32, it implements the steps in the various embodiments of the method for identifying chip unpacking described above, for example... Figure 1 Steps 101 to 104 are shown. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each unit in the above-described device embodiments, for example... Figure 2 The function of the unit shown.

[0091] For example, the computer program 32 can be divided into one or more units, which are stored in the memory 31 and executed by the processor 30 to complete the present invention. The one or more units can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 32 in the terminal device 3. For example, the specific functions of each unit of the computer program 32 can be divided as follows: The acquisition unit is used to acquire the current real-time image and rotate the current real-time image to a preset direction to obtain the current chip image; wherein, the preset direction refers to the direction corresponding to the standard chip area in the standard template image; An acquisition unit is configured to acquire a region of interest in a standard template image, wherein the region of interest includes a chip die region, a wire bonding region, and / or a package pin region. An extraction unit is configured to extract a to-be-recognized region corresponding to the region of interest in the current chip image. A control unit is configured to control a chip unpacking process according to a region similarity between the region of interest and the to-be-recognized region corresponding to the region of interest.

[0092] The terminal device 3 includes, but is not limited to, a processor 30 and a memory 31. Those skilled in the art can understand that, Figure 3 The terminal device 3 is only an example and does not constitute a limitation on the terminal device 3, and can include more or fewer components than shown, or combine certain components, or include different components, for example, the terminal device can also include an input / output device, a network access device, a bus, and the like.

[0093] The processor 30 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0094] The memory 31 can be an internal storage unit of the terminal device 3, for example, a hard disk or a memory of the terminal device 3. The memory 31 can also be an external storage device of the terminal device 3, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 31 can include both the internal storage unit and the external storage device of the terminal device 3. The memory 31 is used to store the computer program and other programs and data required by the terminal device 3. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0095] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiment of the application.

[0096] It should be noted that the information interaction, execution process and the like between the above devices / units are based on the same concept as the method embodiments of the application, and the specific functions and technical effects brought by them can be referred to the method embodiments part. Therefore, no further description is given here.

[0097] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of each functional unit and module is exemplified, and in actual application, the above-mentioned functions can be completed by different functional units and modules according to needs. The internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software functional unit. In addition, the specific name of each functional unit and module is only for convenient distinction, and does not limit the protection scope of the application. The specific working process of the unit and module in the above system can refer to the corresponding process in the above method embodiments, and will not be repeated here.

[0098] The embodiment of the application further provides a computer readable storage medium, the computer readable storage medium stores a computer program, and the computer program is executed by a processor to realize the steps in each method embodiment.

[0099] The embodiment of the application provides a computer program product, when the computer program product is run on a mobile terminal, so that the mobile terminal executes to realize the steps in each method embodiment.

[0100] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the present application can implement all or part of the processes in the above-mentioned embodiment methods, which can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer readable storage medium, and the computer program can implement the steps of each method embodiment when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms. The computer readable medium can at least include any entity or device capable of carrying the computer program code to the photographing device / terminal equipment, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, such as U disk, mobile hard disk, magnetic disk or optical disk.

[0101] In the above embodiments, the description of each embodiment has its own focus, and the parts not described or recorded in detail in a certain embodiment can be referred to the related description of other embodiments.

[0102] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0103] In the embodiments provided by the present application, it should be understood that the disclosed apparatus / network device and method can be implemented in other ways. For example, the above-described apparatus / network device embodiments are merely schematic, for example, the division of the modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0104] The units described as separate components can or can not be physically separate, and the components displayed as separate components can or can not be physical separate, and can be located in one position or distributed on a plurality of network units.

[0105] It should be understood that the term "comprises" or "comprising," when used in this specification and accompanying claims, indicates the presence of the described features, integers, steps, operations, elements, and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0106] It should also be understood that the term "and / or," when used in this specification and accompanying claims, means that one or more of the associated listed items can be present.

[0107] As used in this specification and any claims of this application, the terms "if" and "when" can be construed to mean "upon” or "in response to determining” or “in response to monitoring” depending on the context. Similarly, the phrase “if it is determined” or “if it is monitored that [the described condition or event]” can be construed to mean “upon determining” or “in response to determining” or “upon monitoring [the described condition or event]” or “in response to monitoring [the described condition or event]” depending on the context.

[0108] In addition, the terms "first," "second," "third," etc. are used herein only to describe different instances, and do not imply or suggest relative importance of the described elements.

[0109] The description of the references "one embodiment," "some embodiments," and the like, in the specification, means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. Thus, the appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in additional embodiments," and so on, in various places in the specification are not necessarily all referring to the same embodiment, unless otherwise specified. The terms "comprise," "comprising," "has," "having," "include," "including," and "contains," "containing," or variants thereof, mean "including but not limited to," unless otherwise specified.

[0110] The above-described embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method of identifying the opening of a chip, characterized in that, The method for identifying the opening of the chip comprises: Collecting a current real-time image, and rotating the current real-time image to a preset direction to obtain a current chip image; wherein the preset direction refers to a direction corresponding to a standard chip region in a standard template image; Obtaining a region of interest in the standard template image; wherein the region of interest comprises a chip die region, a wire bonding region and / or a pin region; Extracting a to-be-identified region corresponding to the region of interest in the current chip image; Controlling a chip opening process according to a region similarity between the region of interest and the to-be-identified region corresponding to the region of interest.

2. The method of claim 1, wherein the opening of the chip is identified by the fact that the chip is not in the package. The step of collecting a current real-time image, and rotating the current real-time image to a preset direction to obtain a current chip image comprises: Collecting an unopened chip image, and extracting an unopened chip region in the unopened chip image; Extracting a current identification pattern in the unopened chip region; wherein the current identification pattern refers to a product identification, a logo or a specific pattern printed on a chip packaging region; Extracting a current direction of the current identification pattern; During the chip opening process, collecting a current real-time image, and rotating the current real-time image to a preset direction based on the current direction to obtain a current chip image.

3. The method for identifying chip unpacking as described in claim 2, characterized in that, The step of extracting a current direction of the current identification pattern comprises: Extracting a standard identification pattern in a standard template image; Constructing a first coordinate system with a center of a standard packaging region as an origin and length and width directions as axes; Based on the first coordinate system, extracting a plurality of first sampling points on an edge of the standard identification pattern at a fixed horizontal coordinate interval within a preset angle range; wherein the preset angle range comprises 0 degrees to 90 degrees, 90 degrees to 180 degrees, 180 degrees to 270 degrees or 270 degrees to 360 degrees; Constructing a second coordinate system with a center of the current identification pattern as an origin and length and width directions as axes; Based on the second coordinate system, extracting a plurality of second sampling points on an edge of the current identification pattern at a fixed horizontal coordinate interval within a plurality of current angle ranges; wherein the plurality of current angle ranges comprise 0 degrees to 90 degrees, 90 degrees to 180 degrees, 180 degrees to 270 degrees and 270 degrees to 360 degrees; Calculating a current direction of the current identification pattern according to the plurality of first sampling points and the plurality of second sampling points.

4. The method for identifying chip unpacking as described in claim 3, characterized in that, The step of calculating a current direction of the current identification pattern according to the plurality of first sampling points and the plurality of second sampling points comprises: Calculating a first distance between the plurality of first sampling points and the origin, and constructing the plurality of first distances into a first distance vector; Calculating a second distance between the plurality of second sampling points and the origin, and constructing the plurality of second distances into a second distance vector; Respectively calculating a vector similarity between the second distance vector and the first distance vector corresponding to each of the plurality of current angle ranges; Calculating a direction included angle between a current angle range corresponding to a maximum vector similarity and a preset angle range; Taking the direction included angle as the current direction.

5. The method of claim 1, wherein the opening of the chip is identified by the fact that the chip is not in the package. 5 The step of controlling a chip opening process according to a region similarity between a region of interest and a to-be-identified region corresponding to the region of interest comprises: calculate a region similarity between the region of interest and the region to be identified corresponding to the region of interest; when the region similarity is lower than a preset similarity, continue to execute a chip opening process and collect a subsequent real-time image; when the region similarities are all not lower than the preset similarity, stop the chip opening process.

6. The method of claim 5, wherein the opening of the chip is identified by the fact that the chip is not in the package. 5 The step of calculating the region similarity between the region of interest and the region to be identified corresponding to the region of interest comprises: extracting a first pixel value and a second pixel value corresponding to a same pixel point position in the region of interest and the region to be identified corresponding to the region of interest; calculating an absolute value difference between the first pixel value and the second pixel value; if the absolute value difference is less than a first value, taking the pixel point position as an interested pixel position; if the absolute value difference is not less than the first value, taking the pixel point position as a non-interested pixel position; extracting a neighboring interested pixel position corresponding to each interested pixel position to obtain a current image region composed of multiple neighboring interested pixel positions or a single interested pixel position; counting a number of pixel positions in the current image region; taking a current image region with a number of pixel positions greater than a second number as a target image region; calculating the region similarity according to the number of pixel positions corresponding to the target image region and the absolute value difference corresponding to each pixel point position.

7. The method for identifying chip unpacking as described in claim 6, characterized in that, The step of calculating the region similarity according to the number of pixel positions corresponding to the target image region and the absolute value difference corresponding to each pixel point position comprises: obtaining a maximum absolute value difference; According to the maximum absolute value difference and the absolute value difference corresponding to each pixel point position, a quality weight corresponding to each pixel point position is calculated; wherein the quality weight = a , d i represents the absolute value difference corresponding to the i-th pixel point position, represents the maximum absolute value difference, represents the attenuation coefficient; According to the number of pixel positions corresponding to the target image region and the quality weight corresponding to each pixel position, a region contribution degree is calculated; wherein the region contribution degree , n j represents the number of pixel positions corresponding to the jth target image region, w i represents the quality weight corresponding to the ith pixel position; taking an average value of region contribution degrees corresponding to multiple target image regions as the region similarity.

8. A device for identifying the opening of a chip, characterized in that The device for identifying chip opening comprises: an acquisition unit configured to acquire a current real-time image and rotate the current real-time image to a preset direction to obtain a current chip image, wherein the preset direction refers to a direction corresponding to a standard chip region in a standard template image; an obtaining unit configured to obtain a region of interest in a standard template image, wherein the region of interest comprises a chip die region, a wire bonding region and / or a packaging pin region; an extracting unit configured to extract a region to be identified corresponding to the region of interest in the current chip image; a control unit configured to control a chip opening process according to a region similarity between the region of interest and the region to be identified corresponding to the region of interest.

9. A terminal device, comprising: The terminal device comprises a memory, a processor and a program for identifying chip opening stored in the memory and executable on the processor, and the program for identifying chip opening is configured to implement steps in the method for identifying chip opening according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, the computer-readable storage medium comprising: The computer program is executed by the processor to implement steps in the method for identifying chip opening according to any one of claims 1 to 7.

Images