CN115035944A - Detection method and device of semiconductor chip and computer equipment - Google Patents
Detection method and device of semiconductor chip and computer equipment Download PDFInfo
- Publication number
- CN115035944A CN115035944A CN202210677654.2A CN202210677654A CN115035944A CN 115035944 A CN115035944 A CN 115035944A CN 202210677654 A CN202210677654 A CN 202210677654A CN 115035944 A CN115035944 A CN 115035944A
- Authority
- CN
- China
- Prior art keywords
- image
- target
- semiconductor chip
- similarity
- dimensional code
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
- G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
- G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/56—External testing equipment for static stores, e.g. automatic test equipment [ATE]; Interfaces therefor
- G11C29/56008—Error analysis, representation of errors
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
- G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
- G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
- G11C2029/1206—Location of test circuitry on chip or wafer
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of semiconductor chip detection, and discloses a detection method and a detection device for a semiconductor chip and computer equipment, wherein the method comprises the following steps: acquiring an image of a semiconductor chip as an image to be analyzed; dividing an image to be analyzed into a multi-level image sequence; calculating the similarity between the multi-level image sequence and a preset template image to obtain a target image; comparing the target image with a standard image of a target detection position; when the similarity between the target image and the standard image is lower than a set value, determining that the target detection position has defects; and extracting the two-dimensional code of the semiconductor chip and updating the information in the two-dimensional code of the semiconductor chip. The method, the device and the computer equipment for detecting the semiconductor chip, provided by the invention, can improve the automation level and degree by realizing automatic detection, reduce misjudgment caused by artificial difference and error, improve the reliability of semiconductor chip production, improve the utilization rate of machines and factories and improve the operation cost of enterprises.
Description
Technical Field
The present invention relates to the field of semiconductor chip detection technologies, and in particular, to a method and an apparatus for detecting a semiconductor chip, and a computer device.
Background
At present, the demand of electronic chips is changing day by day along with the rapid development of the electronic information industry, and the insufficient detection method and detection capability of the packaging quality of the chips are common problems in the industry at present. The application of powerful digital image processing technology to the production automation detection of semiconductor packaging detection is the most popular scheme in enterprises at present and is also the demand of actual productivity development instead of the traditional manual detection. In the high development of the semiconductor packaging test industry, the automation degree of modern electronics and machinery is gradually improved, the integration degree of advanced high-tech semiconductor chips is higher and higher, workers rely on limited resources of human eyes for operation, not only the working pressure is high, but also errors are easy to occur in the production and detection processes due to human eye fatigue and human difference, and therefore the urgent requirements of the high-tech industry cannot be met by means of traditional manual quality inspection and detection.
Disclosure of Invention
The invention provides a method and a device for detecting a semiconductor chip and computer equipment, which can improve the automation level and degree by realizing automatic detection, reduce misjudgment caused by artificial difference and error, improve the reliability of semiconductor chip production, improve the utilization rate of machines and factories and improve the operation cost of enterprises.
The invention provides a detection method of a semiconductor chip, which comprises the following steps:
acquiring an image of a semiconductor chip as an image to be analyzed by using a CCD image sensor;
dividing an image to be analyzed into a multi-level image sequence according to a preset template image, and amplifying the multi-level image sequence according to the preset template image;
calculating the similarity between the multi-level image sequence and the preset template image by adopting a sequential correlation algorithm to obtain an image of a target detection position of the semiconductor chip as a target image;
comparing the target image with a standard image of the target detection position;
when the similarity between the target image and the standard image is lower than a set value, correcting the target image, and when the similarity between the corrected target image and the standard image is lower than the set value, determining that the target detection position has a defect;
and extracting the two-dimensional code of the semiconductor chip by using the CCD image sensor, and transmitting the target image through a network so as to update the information in the two-dimensional code of the semiconductor chip.
Further, before the step of acquiring an image of the semiconductor chip as an image to be analyzed by using the CCD image sensor, the method further includes:
identifying a substrate of a semiconductor chip, and printing a two-dimensional code for bearing data information of the semiconductor chip on the substrate by using laser;
extracting the two-dimensional code by using the CCD image sensor, and carrying out data transmission on the data of the semiconductor chip acquired by the CCD image sensor through a network;
the method comprises the steps of obtaining a two-dimensional code image containing semiconductor chip data, carrying out code positioning, separation and decoding on the two-dimensional code image to obtain information in a two-dimensional code, and storing the information in the two-dimensional code in a preset semiconductor chip database.
Further, the step of dividing the image to be analyzed into a multi-level image sequence according to a preset template image includes:
weighting and averaging an image to be analyzed into a primary image formed by one pixel according to the pixel value of every x y pixels of the template image;
dividing the primary image into a plurality of regional images according to preset pixel values;
combining the image to be analyzed, the first-level image and the plurality of area images to form the multi-level image sequence;
sequentially extracting a region image in the multi-level image sequence for expansion to obtain a plurality of secondary region images;
obtaining a row amplification ratio and a column amplification ratio according to a preset template image and a second-level area image;
and obtaining a row mapping value and a column mapping value according to the row amplification ratio and the column amplification ratio, and interpolating the two-level region image according to the row mapping value and the column mapping value to obtain an amplified image of the region image.
Further, after the step of dividing the primary image into a plurality of area images according to preset pixel values, the method further includes:
establishing a rectangular coordinate system by taking the lower left corner of the primary image as an origin;
and obtaining the coordinate point ranges of the plurality of regional images by taking one pixel value as an interval.
Further, the step of calculating the similarity between the multi-level image sequence and the template image by using a sequential correlation algorithm to obtain an image of the target detection position of the semiconductor chip as a target image includes:
extracting an area image of a plurality of area images in the multilevel image sequence as a target area image, and acquiring an amplified image of the target area image;
extracting a plurality of secondary template images with the characteristics of the semiconductor chip in the preset template images according to preset pixel values;
calculating the similarity between the amplified image of the target area image and the plurality of secondary template images, and obtaining a plurality of similarity values;
judging whether the similarity values are larger than a set threshold value or not;
if the similarity values are larger than a set threshold value, the target area image is used as the image of the target detection position;
recording the coordinate point range of the target area image, and extracting a secondary template image with the similarity exceeding a set threshold value with the target area image to be associated with the target area image;
and if the similarity values do not have the similarity value larger than the set threshold value, returning to the step of extracting one area image in the multi-area image as the target area image.
Further, in the step of calculating the similarity between the target area image and the plurality of secondary template images and obtaining a plurality of similarity values, the calculation formula is:
normalizing Q to obtain:
wherein the content of the first and second substances,
is the average of the pixel gray levels of the target area image,
is the average of the pixel gray levels of the template image.
Further, the step of, when the similarity between the target image and the standard image is lower than a set value, correcting the target image, and when the similarity between the corrected target image and the standard image is lower than the set value, determining that the target detection position has a defect includes:
when the similarity between the target image and the standard image is lower than a set value, performing semantic segmentation on the target image to obtain a mask image;
acquiring pixel point characteristics of the mask image and performing principal characteristic analysis to obtain principal characteristic vectors;
determining the rotation angle of the target image according to the main feature vector;
extracting the characteristics of the target image and the corresponding secondary template image, and acquiring the coordinates of the same characteristic endpoint;
overlapping the characteristics of the target image with the characteristics of the corresponding secondary template image to obtain characteristic overlapping points;
calculating a deflection angle according to the feature coincident point and the coordinate of the same feature endpoint;
when the difference value between the rotation angle and the deflection angle is within a set range, rotating the target image according to the rotation angle to obtain a target rotation image;
and when the similarity between the target rotation image and the standard image is lower than a set value, determining that the target detection position has a defect.
The present invention also provides a semiconductor chip inspection apparatus, comprising:
the acquisition module is used for acquiring an image of the semiconductor chip as an image to be analyzed by adopting the CCD image sensor;
the dividing module is used for dividing the image to be analyzed into a multi-level image sequence according to a preset template image;
the calculation module is used for calculating the similarity between the multi-level image sequence and a preset template image by adopting a sequential correlation algorithm so as to obtain an image of a target detection position of the semiconductor chip as a target image;
the comparison module is used for comparing the target image with the standard image of the target detection position;
the determining module is used for determining that the target detection position has defects when the similarity between the target image and the standard image is lower than a set value;
and the updating module is used for extracting the two-dimensional code of the semiconductor chip by using the CCD image sensor and transmitting the target image through a network so as to update the information in the two-dimensional code of the semiconductor chip.
The invention also provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method when executing the computer program.
The invention also provides a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.
The invention has the beneficial effects that:
the CCD image sensor is adopted to collect the image of the semiconductor chip, and then the collected image of the semiconductor chip is divided into a multi-level image sequence, because the refinement degree of the semiconductor chip is high, therefore, the divided multi-level image sequence is amplified, the similarity between the region image in the multi-level image sequence and the template image is calculated in sequence, to find the target detection position image of the semiconductor chip, finally comparing the target detection position image with the standard image, when the similarity between the target image and the standard image is lower than the set value, rotating the target image to determine the similarity again, determining that the target detection position of the semiconductor chip has a defect when the similarity is still lower than a set value, recording the image of the defect position in the two-dimensional code of the semiconductor chip only, therefore, the faulty semiconductor chip can be traced in the following process, and the statistical analysis of multiple faults is facilitated. The automatic detection of the semiconductor chip packaging is realized, the automation level and degree can be greatly improved, the misjudgment caused by artificial difference and error of the current enterprises is greatly reduced, the reliability of the semiconductor chip production can be improved, the utilization rate of machines and factories is improved, and the operation cost of the enterprises is improved.
Drawings
Fig. 1 is a schematic flow chart of a method according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of an apparatus according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of an internal structure of a computer device according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
As shown in fig. 1, the present invention provides a method for inspecting a semiconductor chip, comprising:
s1, acquiring an image of the semiconductor chip as an image to be analyzed by using the CCD image sensor;
s2, dividing an image to be analyzed into a multi-level image sequence according to a preset template image, and amplifying the multi-level image sequence according to the preset template image; (ii) a
S3, calculating the similarity between the multi-level image sequence and the template image by adopting a sequential correlation algorithm to obtain an image of the target detection position of the semiconductor chip as a target image;
s4, comparing the target image with the standard image of the target detection position;
s5, when the similarity between the target image and the standard image is lower than a set value, correcting the target image, and when the similarity between the corrected target image and the standard image is lower than the set value, determining that the target detection position has a defect;
and S6, extracting the two-dimensional code of the semiconductor chip by using the CCD image sensor, and transmitting the target image through a network to update the information in the two-dimensional code of the semiconductor chip.
As described in the foregoing steps S1-S6, the image sensor CCD (charge Coupled device), which is a charge Coupled device, may be generally referred to as a CCD image sensor. It is a semiconductor device which can convert captured optical images into digital signals, has a photoelectric conversion function, and also has functions of storing signal charges, transferring and reading. The image sensor can directly convert the optical signal into an analog current signal, and the amplified current signal is subjected to analog-digital conversion, so that image acquisition, storage, transmission, processing and playback are realized. The CCD image acquisition process mainly converts the shot image from an analog signal into a digital signal which is easy to be identified and distinguished by a computer, and finally converts the digital signal into a digital image, and the image is stored in the local computer and a server.
The CCD image sensor is adopted to collect the image of the semiconductor chip, the collected image of the semiconductor chip is divided into a multi-level image sequence, the divided multi-level image sequence is amplified due to high refinement degree of the semiconductor chip, the similarity between the amplified image of the area image in the multi-level image sequence and the template image is sequentially calculated so as to increase the detection precision, the target detection position image of the semiconductor chip is finally found, the target detection position image is finally compared with the standard image, when the similarity between the target image and the standard image is lower than a set value, the similarity is determined again by rotating the target image due to the influence of the placement angle of the chip on the similarity, when the similarity is still lower than the set value, the target detection position of the semiconductor chip is determined to have a defect, and the image of the defect position is recorded in the two-dimensional code of the semiconductor chip only, therefore, the faulty semiconductor chip can be traced in the following process, and the statistical analysis of multiple faults is facilitated. The automatic detection, automatic tracing, automatic analysis, automatic feedback and automatic alarm of the semiconductor chip packaging are realized, no artificial interference is needed, the automation level and degree are greatly improved, the misjudgment caused by artificial difference and error of the current enterprises is greatly reduced, the reliability of the semiconductor chip production can be improved, the utilization rate of machines and factories is improved, and the operation cost of the enterprises is improved.
In one embodiment, before the step of acquiring the image of the semiconductor chip as the image to be analyzed by using the CCD image sensor, the method further includes:
s01, identifying a substrate of the semiconductor chip, and printing a two-dimensional code for bearing data information of the semiconductor chip on the substrate by using laser;
s02, extracting the two-dimensional code by using the CCD image sensor, and transmitting the data of the semiconductor chip acquired by the CCD image sensor through a network;
s03, acquiring a two-dimensional code image containing semiconductor chip data, performing code positioning, separation and decoding on the two-dimensional code image to obtain information in the two-dimensional code, and storing the information in the two-dimensional code into a preset semiconductor chip database.
As described in the above steps S01-S03, the green paint on the surface of the semiconductor chip substrate is a thin layer of insulating property, and the circuit of the substrate is under the green paint for connecting the chip and the solder ball for communication. For laser marking of semiconductor chip manufacture, one of carbon dioxide laser, semiconductor laser and optical fiber laser can be selected; when carrying out two-dimensional code and printing, still need guarantee copper line circuit layer not exposed, so, can utilize machine laser printing can not influence semiconductor chip's performance on the base plate surface at the chip surface. And then, extracting the two-dimensional code by using a CCD image sensor, and carrying out data transmission on the data of the semiconductor chip acquired by the CCD image sensor through a network so as to add the data information (product information, production time, serial number, logistics information and the like) of the semiconductor chip into the two-dimensional code.
And finally, acquiring a two-dimensional code image containing semiconductor chip data, positioning, separating and decoding the code to obtain information in the two-dimensional code, and storing the information in a preset semiconductor chip database for subsequent tracing and analysis. The code positioning and separation are realized mainly through the following methods and steps: and converting the acquired image containing the two-dimensional code into a binary image by using a threshold value theory of point operation, namely, performing binary mathematical processing on the image of the target. The gray value of the pixel point at the point is set as a threshold value. And performing expansion operation on the obtained binary image, mainly using expansion transformation of mathematical morphology to perform edge detection on the expanded image to obtain the outline of the two-dimensional code area, performing boundary correction, and finally separating out a relatively complete two-dimensional code identification area. In the decoding process, after a standard two-dimensional code image is obtained, grid sampling is carried out on the symbol as early as possible, then the image points on the intersection are sampled, and whether the symbol is a dark color block or a light color block is determined according to the current threshold value. Constructing a reasonable bitmap, using binary 1 to represent deeper pixel, using 0 to represent relatively shallower pixel point, thus obtaining the most original binary value of the two-dimensional code, then adopting data queue to these data, and aiming at the wrong value, using the coding rule of the two-dimensional code logic to convert these original data bit streams into data code words, thus realizing decoding.
The database of the two-dimension code corresponding to each semiconductor chip is established to record the production history of each semiconductor chip in real time for tracing, so that enterprises can be helped to collect engineering data and help production to quickly find out the chip fault, and then feedback production is improved.
In one embodiment, the step of dividing an image to be analyzed into a multi-level image sequence according to a preset template image and amplifying the multi-level image sequence according to the preset template image includes:
s21, averaging the image to be analyzed into a primary image formed by one pixel according to the pixel value of each x y pixels of the template image in a weighted mode;
s22, dividing the primary image into a plurality of regional images according to preset pixel values;
s23, combining the image to be analyzed, the primary image and the plurality of regional images to form the multi-level image sequence;
s24, sequentially extracting a region image in the multi-level image sequence for expansion to obtain a plurality of secondary region images;
s25, obtaining a line amplification scale and a column amplification scale according to a preset template image and a preset area image;
and S26, obtaining a row mapping value and a column mapping value according to the row amplification ratio and the column amplification ratio, and interpolating the two-level region image according to the row mapping value and the column mapping value to obtain an amplified image of the region image.
As described in the above steps S21-S26, the image to be analyzed is weighted and averaged into a first-level image composed of one pixel according to the pixel value of every x y pixels of the template image, that is, the image to be analyzed and the template image have the same pixel, then the first-level image is divided into a plurality of area images according to the preset pixel value, when the area images are divided, the area images have continuity, when the side of one line or one column of the image is less than the preset pixel value, the image to be analyzed is determined as an area image, so a plurality of area images are obtained, and finally the image to be analyzed, the first-level image and the plurality of area images are combined to form the multi-level image sequence, so as to extract the images for similarity calculation. Constructing an expanded matrix, and copying pixel values of a first row of an area image to the middle position of the first row of the expanded matrix; copying pixel values of a first row and a first column of the regional image to a position of the first row and the first column of the expanded matrix; copying the pixel value of the last column of the first row of the regional image to the position of the last column of the first row of the expanded matrix to form the pixel value of the first row of the expanded matrix; obtaining the pixel values of the last row, the first column and the last column of the expanded matrix according to the obtaining mode of the pixel value of the first row of the expanded matrix; and copying the pixel values of the area image to the middle position of the expanded matrix, and finally obtaining a secondary area image. And calculating a line amplification ratio and a column amplification ratio according to a preset template image and a two-level area image, wherein the area image is a b, the template image is M N (M > a, N > b), the amplification ratio in the line direction is calculated to be x-N/b, and the amplification ratio in the column direction is calculated to be y-M/a. And calculating the mapping value in the row direction to be 1/x according to the amplification ratio x in the row direction, calculating the mapping value in the column direction to be 1/y according to the amplification ratio y in the column direction, and performing interpolation in the column direction and the row direction on the two-level region image by using the mapping value by adopting the prior art to obtain an amplified image of the region image.
In one embodiment, after the step of dividing the primary image into a plurality of area images according to preset pixel values, the method further includes:
s221, establishing a rectangular coordinate system by taking the lower left corner of the primary image as an origin;
s222, taking a pixel value as an interval, and obtaining coordinate point ranges of the plurality of regional images.
As described in the foregoing steps S221 to S222, a rectangular coordinate system is established with the lower left corner of the primary image as the origin, then a value on a coordinate axis is obtained with a pixel value as an interval, that is, one pixel on the coordinate axis is 1, two pixels are 2 …, and so on in the rectangular coordinate system, and a coordinate point range of a plurality of area images can be obtained according to the value on the coordinate axis, so as to record the position of the required area image in the primary image, that is, the position of the required detection position in the semiconductor chip is obtained.
In one embodiment, the step of calculating the similarity between the multi-level image sequence and a preset template image by using a sequential correlation algorithm to obtain an image of a target detection position of the semiconductor chip as a target image includes:
s31, extracting an area image of a plurality of area images in the multilevel image sequence as a target area image, and acquiring an enlarged image of the target area image;
s32, extracting a plurality of secondary template images with the semiconductor chip characteristics in the preset template images according to preset pixel values;
s33, calculating the similarity between the amplified image of the target area image and the plurality of secondary template images, and obtaining a plurality of similarity values;
s34, judging whether the similarity values are larger than a set threshold value or not;
s35, if there is a similarity value greater than a set threshold among the plurality of similarity values, taking the target area image as the image of the target detection position;
s36, recording the coordinate point range of the target area image, and extracting a secondary template image with the similarity exceeding a set threshold value with the target area image to be associated with the target area image;
s37, if none of the similarity values is greater than the set threshold, the process returns to the step of extracting one of the multi-region images as the target region image.
As described in the above steps S31-S35, one area image of the several area images in the multi-level image sequence is sequentially extracted as a target area image, and an enlarged image of the target area image is obtained so as to analyze the several areas of the primary image one by one; extracting a plurality of secondary template images with semiconductor chip characteristics in the preset template images according to the preset pixel values, wherein each secondary template image is an image of the semiconductor chip characteristics and is also a template image corresponding to a fault point of a subsequent semiconductor chip; calculating the similarity between the amplified image of the target area image and the plurality of secondary template images, and obtaining a plurality of similarity values, wherein the similarity between each target area image and each secondary template image is calculated because the corresponding relation between the target area image and the plurality of secondary template images is unknown, for example, the similarity between each target area image and each secondary template image is 100, and the similarity between each two secondary template images is 100 × 500 if 50 secondary template images are obtained; judging whether the similarity values are greater than a set threshold value or not, wherein 50 similarity values are obtained by 1 target area and 50 secondary template images, judging whether the similarity values are greater than the set threshold value or not, if so, indicating that the target area image and the corresponding secondary template image have a corresponding association relationship and representing that the target area is an image of a target detection position, recording the coordinate point range of the target area image, extracting the secondary template image with the similarity greater than the set threshold value with the target area image and associating the secondary template image with the target area image so as to trace the target detection position subsequently; if the similarity value larger than the set threshold value is not contained in the 50 similarity values, the target area image is not the image of the target detection position, the step of extracting one area image in the multi-area image as the target area image is returned, and the similarity calculation of the next target area image is carried out.
In one embodiment, in the step of calculating the similarity between the target area image and each of the plurality of template images and obtaining a plurality of similarity values, the calculation formula is:
normalizing Q yields:
wherein the content of the first and second substances,
is the average of the pixel gray levels of the target area image,
is the average of the pixel gray levels of the template image.
In one embodiment, the step of, when the similarity between the target image and the standard image is lower than a set value, correcting the target image, and when the similarity between the corrected target image and the standard image is lower than the set value, determining that the target detection position has a defect includes:
s51, when the similarity between the target image and the standard image is lower than a set value, performing semantic segmentation on the target image to obtain a mask image;
s52, obtaining pixel point characteristics of the mask image and performing main characteristic analysis to obtain a main characteristic vector;
s53, determining the rotation angle of the target image according to the main feature vector;
s54, extracting the characteristics of the target image and the corresponding secondary template image, and acquiring the coordinates of the same characteristic endpoint;
s55, overlapping the characteristics of the target image and the characteristics of the corresponding secondary template image to obtain characteristic overlapping points;
s56, calculating a deflection angle according to the feature coincident point and the coordinate of the same feature endpoint;
s57, when the difference value between the rotation angle and the deflection angle is within a set range, rotating the target image according to the rotation angle to obtain a target rotation image;
and S58, when the similarity between the target rotation image and the standard image is lower than a set value, determining that the target detection position has a defect.
As described in the above steps S51-S58, since the angle of the image also affects the similarity of the image, when the similarity between the target image and the standard image is lower than the set value, it is necessary to determine whether the angle of the image affects the similarity of the image. Performing Semantic Segmentation (Semantic Segmentation) algorithm on a target image, such as FCNN, segNet, DeepLab v1/v2/v3, watercut, grabcut to obtain a Mask image (Mask), acquiring pixel point characteristics of the Mask image, performing main characteristic analysis to obtain a main characteristic vector, converting a group of variables possibly having correlation into a group of linear uncorrelated variables through orthogonal transformation in the main characteristic analysis, and converting the group of variables into the main characteristic vector. And finally, acquiring a reference direction according to the main characteristic vector, calculating an included angle between the main characteristic vector and the reference direction, and further determining the rotation angle of the target image according to the included angle. Extracting feature coordinates of a target image and a corresponding secondary template image, obtaining coordinates of the same feature end point, namely the coordinates of the same position of the target image and the secondary template image, then overlapping the features of the target image and the features of the corresponding secondary template image to obtain feature overlapping coordinates, calculating the deflection angle of the feature end point according to the overlapping coordinates and the coordinates of the two feature end points, calculating the angle of the target image in multiple directions to avoid errors, and finally when the difference value of the rotation angle and the deflection angle is within a set range, indicating that the rotation angle can be adopted for rotating, so that the target image is rotated according to the rotation angle to obtain a target rotation image; and calculating the similarity between the target rotation image and the standard image again, and determining that the target detection position has defects when the similarity between the target rotation image and the standard image is lower than a set value.
In one embodiment, before the step of extracting the two-dimensional code of the semiconductor chip by using the CCD image sensor and transmitting the target image through a network to update information in the two-dimensional code of the semiconductor chip, the method further includes:
s061, adding a fault mark convenient for follow-up tracing to the target image, and starting an alarm device to give an alarm to remind a manager that the current semiconductor chip is detected to be faulty.
As described in the step S061, the system automatically compares the target image with the template image, and determines and marks the failure according to the standard, and records the two-dimensional code information, the engineering data, and the corresponding image information of the failure of the semiconductor chip, so as to facilitate the tracing of the semiconductor chip, where the tracing of the semiconductor chip is to give a unique identification to each chip product, i.e., the two-dimensional code, and all products circulating in the market can know the material composition, production process, sales channel, and core test performance index of the semiconductor chip through the two-dimensional code. The application of the two-dimensional code is utilized, a perfect semiconductor chip database system is established by combining the acquisition of image processing, the quality index of the whole production flow is monitored and analyzed in real time, the quick automatic response of partial engineering problems is replaced, the product tracing and tracing are realized, the influence range of the problem products and the product recovery of the products sold out are reduced, the economic loss and the mental loss of consumers and enterprises are reduced, and the limited manual labor force is greatly liberated.
As shown in fig. 2, the present invention also provides a semiconductor chip inspection apparatus, comprising:
the acquisition module 1 is used for acquiring an image of a semiconductor chip as an image to be analyzed by adopting a CCD image sensor;
the dividing module 2 is used for dividing an image to be analyzed into a multi-level image sequence according to a preset template image and amplifying the multi-level image sequence according to the preset template image;
the calculation module 3 is used for calculating the similarity between the multi-level image sequence and a preset template image by adopting a sequential correlation algorithm so as to obtain an image of a target detection position of the semiconductor chip as a target image;
a comparison module 4, configured to compare the target image with a standard image of the target detection position;
a determining module 5, configured to correct the target image when the similarity between the target image and the standard image is lower than a set value, and determine that a defect exists at the target detection position when the similarity between the corrected target image and the standard image is lower than the set value;
and the updating module 6 is used for extracting the two-dimensional code of the semiconductor chip by using the CCD image sensor and transmitting the target image through a network so as to update the information in the two-dimensional code of the semiconductor chip.
In one embodiment, further comprising:
the identification module is used for identifying a substrate of a semiconductor chip and printing a two-dimensional code for bearing data information of the semiconductor chip on the substrate by utilizing laser;
the transmission module is used for extracting the two-dimensional code by using the CCD image sensor and transmitting the data of the semiconductor chip acquired by the CCD image sensor through a network;
the storage module is used for acquiring a two-dimensional code image containing semiconductor chip data, performing code positioning, separation and decoding on the two-dimensional code image to obtain information in the two-dimensional code, and storing the information in the two-dimensional code into a preset semiconductor chip database.
In one embodiment, the partitioning module 2 includes:
the primary image unit is used for weighting and averaging the image to be analyzed into a primary image formed by one pixel according to the pixel value of each x y pixels of the template image;
the regional image unit is used for dividing the primary image into a plurality of regional images according to preset pixel values;
the combination unit is used for combining the image to be analyzed, the first-level image and the plurality of area images to form the multi-level image sequence;
the second-level area image unit is used for sequentially extracting an area image in the multi-level image sequence to be expanded to obtain a plurality of second-level area images;
the magnification ratio unit is used for obtaining a row magnification ratio and a column magnification ratio according to a preset template image and a two-stage area image;
and the amplified image unit is used for obtaining a row mapping value and a column mapping value according to the row amplification ratio and the column amplification ratio, and interpolating the two-level region image according to the row mapping value and the column mapping value to obtain an amplified image of the region image.
In one embodiment, further comprising:
the coordinate system module is used for establishing a rectangular coordinate system by taking the lower left corner of the primary image as an origin;
and the coordinate point range module is used for obtaining the coordinate point ranges of the plurality of regional images by taking one pixel value as an interval.
In one embodiment, the calculation module 3 includes:
the target area image extracting unit is used for extracting an area image of a plurality of area images in the multilevel image sequence as a target area image and acquiring an amplified image of the target area image;
the second-level template image unit is used for extracting a plurality of second-level template images with the characteristics of the semiconductor chip in the preset template images according to preset pixel values;
the similarity value calculation unit is used for calculating the similarity between the amplified image of the target area image and the plurality of secondary template images and obtaining a plurality of similarity values;
a judging unit configured to judge whether or not there is a similarity value larger than a set threshold value among the plurality of similarity values;
a target detection position unit configured to take the target area image as an image of the target detection position when a similarity value larger than a set threshold value is included among a plurality of similarity values;
the association unit is used for recording the coordinate point range of the target area image, and extracting a secondary template image with similarity exceeding a set threshold value with the target area image to be associated with the target area image;
and a returning unit for returning to the step of extracting one of the multi-region images as the target region image when none of the plurality of similarity values has a similarity value greater than a set threshold.
In one embodiment, in the similarity value calculation unit, the calculation formula is:
normalizing Q yields:
wherein the content of the first and second substances,
is the average of the pixel gray levels of the target area image,
is the average of the pixel gray levels of the template image.
In one embodiment, the determining module 6 includes:
the mask image unit is used for performing semantic segmentation on the target image to obtain a mask image when the similarity between the target image and the standard image is lower than a set value;
the main characteristic analysis unit is used for acquiring the pixel point characteristics of the mask image and performing main characteristic analysis to obtain a main characteristic vector;
a rotation angle unit, configured to determine a rotation angle of the target image according to the principal feature vector;
the characteristic end point unit is used for extracting the characteristics of the target image and the corresponding secondary template image and acquiring the coordinates of the same characteristic end point;
the characteristic coincidence point unit is used for coinciding the characteristics of the target image with the characteristics of the corresponding secondary template image to obtain characteristic coincidence points;
the deflection angle unit is used for calculating a deflection angle according to the feature coincident point and the coordinate of the same feature endpoint;
the target rotating image unit is used for rotating the target image according to the rotating angle to obtain a target rotating image when the difference value between the rotating angle and the deflection angle is within a set range;
and the defect unit is used for determining that the target detection position has defects when the similarity between the target rotation image and the standard image is lower than a set value.
In one embodiment, further comprising:
and the marking module is used for adding a fault mark convenient for follow-up tracing for the target image and starting the alarm device to give an alarm to remind a manager that the current semiconductor chip is detected to have a fault.
The above modules and units are all used for correspondingly executing each step in the method for detecting a semiconductor chip, and the specific implementation manner thereof is described with reference to the above method embodiment, and is not described herein again.
As shown in fig. 3, the present invention also provides a computer device, which may be a server, and the internal structure of which may be as shown in fig. 3. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer designed processor is used to provide computational and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the operating system and the running of computer programs in the non-volatile storage medium. The database of the computer device is used to store all data required by the process of the inspection method of the semiconductor chip. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of testing a semiconductor chip.
It will be understood by those skilled in the art that the structure shown in fig. 3 is only a block diagram of a part of the structure related to the present application, and does not constitute a limitation to the computer device to which the present application is applied.
An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the method for detecting any one of the semiconductor chips.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by hardware related to instructions of a computer program, which may be stored in a non-volatile computer-readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium provided herein and used in the examples may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double-rate SDRAM (SSRSDRAM), Enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A method for inspecting a semiconductor chip, comprising:
acquiring an image of a semiconductor chip as an image to be analyzed by adopting a CCD image sensor;
dividing an image to be analyzed into a multi-level image sequence according to a preset template image, and amplifying the multi-level image sequence according to the preset template image;
calculating the similarity between the multi-level image sequence and the preset template image by adopting a sequential correlation algorithm to obtain an image of a target detection position of the semiconductor chip as a target image;
comparing the target image with a standard image of the target detection position;
when the similarity between the target image and the standard image is lower than a set value, correcting the target image, and when the similarity between the corrected target image and the standard image is lower than the set value, determining that the target detection position has a defect;
and extracting the two-dimensional code of the semiconductor chip by using the CCD image sensor, and transmitting the target image through a network so as to update the information in the two-dimensional code of the semiconductor chip.
2. The method for inspecting a semiconductor chip according to claim 1, wherein the step of acquiring the image of the semiconductor chip as the image to be analyzed using the CCD image sensor further comprises:
identifying a substrate of a semiconductor chip, and printing a two-dimensional code for bearing data information of the semiconductor chip on the substrate by using laser;
extracting the two-dimensional code by using the CCD image sensor, and carrying out data transmission on the data of the semiconductor chip acquired by the CCD image sensor through a network;
the method comprises the steps of obtaining a two-dimensional code image containing semiconductor chip data, carrying out code positioning, separation and decoding on the two-dimensional code image to obtain information in a two-dimensional code, and storing the information in the two-dimensional code in a preset semiconductor chip database.
3. The method of claim 1, wherein the step of dividing the image to be analyzed into a multi-level image sequence according to a preset template image and magnifying the multi-level image sequence according to the preset template image comprises:
weighting and averaging an image to be analyzed into a primary image formed by one pixel according to the pixel value of every x y pixels of the template image;
dividing the primary image into a plurality of regional images according to preset pixel values;
combining the image to be analyzed, the first-level image and the plurality of area images to form the multi-level image sequence;
sequentially extracting a region image in the multilevel image sequence for expansion to obtain a plurality of secondary region images;
obtaining a row amplification ratio and a column amplification ratio according to a preset template image and a second-level area image;
and obtaining a row mapping value and a column mapping value according to the row amplification ratio and the column amplification ratio, and interpolating the two-level region image according to the row mapping value and the column mapping value to obtain an amplified image of the region image.
4. The method for inspecting a semiconductor chip according to claim 3, wherein the step of dividing the primary image into a plurality of area images according to the predetermined pixel values is followed by the step of:
establishing a rectangular coordinate system by taking the lower left corner of the primary image as an origin;
and obtaining the coordinate point ranges of the plurality of regional images by taking one pixel value as an interval.
5. The method according to claim 4, wherein the step of calculating the similarity between the multi-level image sequence and the template image by using a sequential correlation algorithm to obtain an image of the target detection position of the semiconductor chip as the target image comprises:
extracting one area image of a plurality of area images in the multilevel image sequence as a target area image, and acquiring an amplified image of the target area image;
extracting a plurality of secondary template images with the characteristics of the semiconductor chip in the preset template images according to preset pixel values;
calculating the similarity between the amplified image of the target area image and the plurality of secondary template images, and obtaining a plurality of similarity values;
judging whether the similarity values are larger than a set threshold value or not;
if the similarity values are larger than a set threshold value, the target area image is used as the image of the target detection position;
recording the coordinate point range of the target area image, and extracting a secondary template image with the similarity exceeding a set threshold value with the target area image to be associated with the target area image;
and if the similarity values do not have the similarity value larger than the set threshold value, returning to the step of extracting one area image in the multi-area image as the target area image.
6. The method of claim 5, wherein the step of calculating the similarity between the target area image and the plurality of secondary template images to obtain a plurality of similarity values comprises the following formula:
normalizing Q yields:
wherein the content of the first and second substances,
is the average of the pixel gray levels of the target area image,
is the average of the pixel gray levels of the template image.
7. The method of claim 1, wherein the step of correcting the target image when the similarity between the target image and the standard image is lower than a set value, and determining that the target inspection position has a defect when the similarity between the corrected target image and the standard image is lower than a set value comprises:
when the similarity between the target image and the standard image is lower than a set value, performing semantic segmentation on the target image to obtain a mask image;
acquiring pixel point characteristics of the mask image and performing main characteristic analysis to obtain a main characteristic vector;
determining the rotation angle of the target image according to the main feature vector;
extracting the characteristics of the target image and the corresponding secondary template image, and acquiring the coordinates of the same characteristic endpoint;
overlapping the characteristics of the target image with the characteristics of the corresponding secondary template image to obtain characteristic overlapping points;
calculating a deflection angle according to the feature coincident point and the coordinate of the same feature endpoint;
when the difference value between the rotation angle and the deflection angle is within a set range, rotating the target image according to the rotation angle to obtain a target rotation image;
and when the similarity between the target rotating image and the standard image is lower than a set value, determining that the target detection position has a defect.
8. An inspection apparatus for a semiconductor chip, comprising:
the acquisition module is used for acquiring an image of the semiconductor chip as an image to be analyzed by adopting the CCD image sensor;
the device comprises a dividing module, a judging module and a processing module, wherein the dividing module is used for dividing an image to be analyzed into a multi-level image sequence according to a preset template image and amplifying the multi-level image sequence according to the preset template image;
the calculation module is used for calculating the similarity between the multi-level image sequence and a preset template image by adopting a sequential correlation algorithm so as to obtain an image of a target detection position of the semiconductor chip as a target image;
the comparison module is used for comparing the target image with the standard image of the target detection position;
the determining module is used for correcting the target image when the similarity between the target image and the standard image is lower than a set value, and determining that the target detection position has defects when the similarity between the corrected target image and the standard image is lower than the set value;
and the updating module is used for extracting the two-dimensional code of the semiconductor chip by using the CCD image sensor and transmitting the target image through a network so as to update the information in the two-dimensional code of the semiconductor chip.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210677654.2A CN115035944A (en) | 2022-06-15 | 2022-06-15 | Detection method and device of semiconductor chip and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210677654.2A CN115035944A (en) | 2022-06-15 | 2022-06-15 | Detection method and device of semiconductor chip and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115035944A true CN115035944A (en) | 2022-09-09 |
Family
ID=83124795
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210677654.2A Pending CN115035944A (en) | 2022-06-15 | 2022-06-15 | Detection method and device of semiconductor chip and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115035944A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116051564A (en) * | 2023-04-02 | 2023-05-02 | 广东仁懋电子有限公司 | Chip packaging defect detection method and system |
| CN116402820A (en) * | 2023-06-07 | 2023-07-07 | 合肥联宝信息技术有限公司 | Detection method, detection device, detection equipment and storage medium |
| CN117074845A (en) * | 2023-10-16 | 2023-11-17 | 广州市零脉信息科技有限公司 | Electronic component quality detection platform |
-
2022
- 2022-06-15 CN CN202210677654.2A patent/CN115035944A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116051564A (en) * | 2023-04-02 | 2023-05-02 | 广东仁懋电子有限公司 | Chip packaging defect detection method and system |
| CN116402820A (en) * | 2023-06-07 | 2023-07-07 | 合肥联宝信息技术有限公司 | Detection method, detection device, detection equipment and storage medium |
| CN117074845A (en) * | 2023-10-16 | 2023-11-17 | 广州市零脉信息科技有限公司 | Electronic component quality detection platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115035944A (en) | Detection method and device of semiconductor chip and computer equipment | |
| CN111612763B (en) | Mobile phone screen defect detection method, device and system, computer equipment and medium | |
| CN102663360B (en) | Method for automatic identifying steel slab coding and steel slab tracking system | |
| CN112766274A (en) | Water gauge image water level automatic reading method and system based on Mask RCNN algorithm | |
| CN113344857B (en) | Defect detection network training method, defect detection method and storage medium | |
| TW202239281A (en) | Electronic substrate defect detection | |
| CN111862057B (en) | Picture labeling method and device, sensor quality detection method and electronic equipment | |
| CN113255590A (en) | Defect detection model training method, defect detection method, device and system | |
| CN111563896A (en) | Image processing method for catenary anomaly detection | |
| CN114511718A (en) | Intelligent management method and system for materials for building construction | |
| CN116681979A (en) | Power equipment target detection method under complex environment | |
| CN115775246A (en) | Method for detecting defects of PCB (printed circuit board) components | |
| CN114267029A (en) | Lane line detection method, device, equipment and storage medium | |
| CN107480580A (en) | Image-recognizing method and pattern recognition device | |
| CN113034492A (en) | Printing quality defect detection method and storage medium | |
| CN115880683B (en) | Urban waterlogging ponding intelligent water level detection method based on deep learning | |
| CN111291712A (en) | Forest fire recognition method and device based on interpolation CN and capsule network | |
| CN114049569B (en) | Deep learning model performance evaluation method and system | |
| CN113298755B (en) | Method and device for rapidly detecting ecological environment change patch based on time sequence image | |
| CN113554018B (en) | Meter automatic modeling method based on machine vision | |
| CN117522867B (en) | Display screen abnormal color detection method based on image characteristics | |
| CN113706620B (en) | Positioning method, positioning device and movable platform based on reference object | |
| CN117474959B (en) | Target object motion trail processing method and system based on video data | |
| CN112541910B (en) | End face gap detection method, device, equipment and medium based on deep learning | |
| Hassan et al. | Analog to Digital Meter Reader Converter Using Signal Processing Technique |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |