A bare chip screening method, device, equipment and storage medium

Automatically screening of naked films through image recognition technology solves the problem of low manual screening efficiency and improves the utilization rate of defective products.

CN114798479BActive Publication Date: 2024-12-03HOSIN GLOBAL ELECTRONICS CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202210232772.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-09
Publication Date
2024-12-03
Estimated Expiration
2042-03-09

AI Technical Summary

Technical Problem

In the prior art, naked film screening mainly relies on manual visual inspection, resulting in low efficiency and low utilization rate of defective products.

Method used

By obtaining the naked chip distribution image of the wafer to be screened, establishing a coordinate system, obtaining the coordinate value of the naked chip, and using artificial intelligence models to identify images, judging the appearance integrity of the naked chip, and realizing automatic filtering.

Benefits of technology

It improves the efficiency of naked-film screening, reduces labor costs, and improves the utilization rate of bad products by marking the retention of bad products.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114798479B_ABST
    Figure CN114798479B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of semiconductor technology, and discloses a method, device, equipment and storage medium for screening bare chips. The screening method comprises: obtaining a first image in a wafer to be screened, wherein the first image is a bare chip distribution image in the wafer to be screened; establishing a coordinate system based on the first image, and obtaining the coordinate value of each bare chip in the wafer to be screened according to the coordinate system; obtaining a second image in the wafer to be screened based on each coordinate value, wherein the second image is a bare chip image in the wafer to be screened; identifying the second image, and judging whether the second image is a bare chip image with a complete appearance, and obtaining a judgment result; according to the judgment result, determining whether the bare chip corresponding to the second image is complete, and performing a screening operation on the bare chip. The intelligent bare chip screening is realized, and the screening efficiency, the screening accuracy and the utilization rate of defective products are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of semiconductor technology, and in particular to a bare chip screening method, device, equipment and storage medium. Background Art

[0002] In the field of semiconductor manufacturing, due to the crystal defects, shape limitations, mechanical damage caused by cutting and other operations of the wafer itself, some of the bare chips will also have different defects or damage in appearance. Therefore, it is necessary to screen out defective bare chips to improve the yield rate of bare chips. The wafer processing process and wafer probe test process are mainly for the production of bare chips before packaging. In the wafer processing process, several layers of circuits are completed on the wafer through epitaxy-oxidation-CVD-sputtering-photolithography-etching-ion implantation and other operations, and multiple bare chips are formed by cutting the wafer; the wafer probe test process is the situation where products of different types or specifications grow on one wafer, and the electrical characteristics of each bare chip are tested by the probe tester, and the bare chips are classified according to different electrical characteristics. At present, the classification of bare chips is basically carried out by employees for visual inspection, picking out the normal bare chips in the wafer and discarding the remaining defective bare chips. Summary of the invention

[0003] The main purpose of the present invention is to provide a bare chip screening method, device, equipment and storage medium, aiming to solve the technical problem that manual bare chip screening consumes a lot of manpower and time, resulting in low efficiency and low utilization rate of defective products.

[0004] To achieve the above object, the present invention provides a first aspect of a bare chip screening method, the screening method comprising:

[0005] Acquire a first image of the wafer to be screened, wherein the first image is a distribution image of bare dies in the wafer to be screened;

[0006] Based on the first image, a coordinate system is established, and coordinate values ​​of each die in the wafer to be screened are obtained according to the coordinate system;

[0007] Based on each of the coordinate values, obtaining a second image of the wafer to be screened, wherein the second image is an image of a bare die in the wafer to be screened;

[0008] Performing image recognition on the second image, and determining whether the second image is a bare die image with a complete appearance, to obtain a determination result;

[0009] According to the judgment result, it is determined whether the die corresponding to the second image is complete, and a screening operation is performed on the die.

[0010] Optionally, in a first implementation of the first aspect of the present invention, the step of acquiring a second image of the wafer to be screened based on each of the coordinate values, wherein the second image is an image of a bare die in the wafer to be screened, includes:

[0011] Based on each of the coordinate values, a photographing device is controlled to photograph an image of a die in the wafer to be screened that corresponds to the coordinate value, so as to obtain the die image.

[0012] Optionally, in a second implementation of the first aspect of the present invention, the step of acquiring a second image of the wafer to be screened based on each of the coordinate values, wherein the second image is an image of a bare die in the wafer to be screened, includes:

[0013] According to a preset segmentation rule, segmenting the first image into a plurality of the second images;

[0014] Based on each of the coordinate values, a second image corresponding to the coordinate value is acquired.

[0015] Optionally, in a third implementation manner of the first aspect of the present invention, before performing image recognition on the second image and determining whether the second image is a bare die image with a complete appearance and obtaining a determination result, the method further includes:

[0016] The second image is input into a preset artificial intelligence model.

[0017] Optionally, in a fourth implementation of the first aspect of the present invention, the judgment result includes:

[0018] The second image is an image of a die with a complete appearance; or, the second image is an image of a die with an incomplete appearance.

[0019] Optionally, in a fifth implementation manner of the first aspect of the present invention, determining whether the die corresponding to the second image is complete according to the judgment result, and performing a screening operation on the die includes:

[0020] If the second image is an image of a die with a complete appearance, determining that the die corresponding to the second image is complete, and placing the die in the first area;

[0021] If the second image is an image of a die with an incomplete appearance, it is determined that the die corresponding to the second image is incomplete, and the die is marked.

[0022] Optionally, in a sixth implementation method of the first aspect of the present invention, the step of establishing a coordinate system based on the first image includes: taking the center of the wafer to be screened as the origin of the coordinate system, and establishing a two-dimensional coordinate system along the horizontal and vertical directions of the plane of the wafer to be screened.

[0023] Optionally, in a seventh implementation manner of the first aspect of the present invention, after establishing a coordinate system based on the first image and acquiring coordinate values ​​of each die in the wafer to be screened according to the coordinate system, the method further includes:

[0024] Based on the coordinate system and the first image, the wafer to be screened is divided into a central area and an outer edge area from the origin to the outside.

[0025] Optionally, in an eighth implementation of the first aspect of the present invention, after dividing the wafer to be screened into a central area and an outer edge area from the origin to the outside in sequence based on the coordinate system and the first image, it further includes:

[0026] Based on the coordinate values, determining the region where each of the bare dies in the wafer to be screened is located;

[0027] If the die is located in the central area and the die has been marked, marking a plurality of adjacent dies of the die;

[0028] If the die is located in the outer edge area, the die is marked.

[0029] Optionally, in a ninth implementation manner of the first aspect of the present invention, after marking the die if the die is located in the outer edge area, the method further includes:

[0030] Acquire a third image of the wafer to be screened, wherein the third image is a circuit image of the marked bare die;

[0031] Performing image recognition on the third image, and determining whether the third image is a qualified circuit image;

[0032] If the third image is a qualified circuit image, placing the bare die in the second area;

[0033] If the third image is an unqualified circuit image, the bare die is placed in a third area.

[0034] A second aspect of the present invention provides a bare chip screening device, the screening device comprising:

[0035] A first acquisition module is used to acquire a first image of the wafer to be screened, wherein the first image is a distribution image of bare dies in the wafer to be screened;

[0036] An establishing module, used to establish a coordinate system based on the first image, and obtain the coordinate value of each bare die in the wafer to be screened according to the coordinate system;

[0037] A second acquisition module, configured to acquire a second image of the wafer to be screened based on each of the coordinate values, wherein the second image is an image of a bare die in the wafer to be screened;

[0038] A first recognition module is used to perform image recognition on the second image, and determine whether the second image is a bare die image with a complete appearance, and obtain a determination result;

[0039] A screening module is used to determine whether the die corresponding to the second image is complete according to the judgment result, and perform a screening operation on the die.

[0040] Optionally, in a first implementation of the second aspect of the present invention, the bare chip screening device further includes:

[0041] An illumination module is used to illuminate the die to be photographed.

[0042] Optionally, in a second implementation of the second aspect of the present invention, the step of acquiring a second image of the wafer to be screened based on each of the coordinate values, wherein the second image is an image of a bare die in the wafer to be screened, includes:

[0043] The photographing unit is used to control the photographing device to photograph the image of the bare die corresponding to the coordinate value in the wafer to be screened based on each of the coordinate values ​​to obtain the bare die image.

[0044] Optionally, in a third implementation of the second aspect of the present invention, the step of acquiring a second image of the wafer to be screened based on each of the coordinate values, wherein the second image is an image of a bare die in the wafer to be screened, includes:

[0045] a segmentation unit, configured to segment the first image into a plurality of the second images according to a preset segmentation rule;

[0046] The first acquisition unit is used to acquire a second image corresponding to the coordinate value based on each of the coordinate values.

[0047] Optionally, in a fourth implementation manner of the second aspect of the present invention, before performing image recognition on the second image and determining whether the second image is a bare die image with a complete appearance, and obtaining a determination result, the method further includes:

[0048] An input module is used to input the second image into a preset artificial intelligence model.

[0049] Optionally, in a fifth implementation of the second aspect of the present invention, the judgment result includes:

[0050] The second image is an image of a die with a complete appearance; or, the second image is an image of a die with an incomplete appearance.

[0051] Optionally, in a sixth implementation manner of the second aspect of the present invention, determining whether the die corresponding to the second image is complete according to the judgment result, and performing a screening operation on the die includes:

[0052] a first determining unit, configured to determine that the die corresponding to the second image is complete if the second image is an image of a die with a complete appearance, and place the die in a first area;

[0053] A marking unit is configured to determine that the die corresponding to the second image is incomplete and mark the die if the second image is an image of a die with an incomplete appearance.

[0054] Optionally, in a seventh implementation method of the second aspect of the present invention, the step of establishing a coordinate system based on the first image includes: establishing a unit for establishing a two-dimensional coordinate system along the horizontal and vertical directions of the plane of the wafer to be screened by taking the center of the circle of the wafer to be screened as the origin of the coordinate system.

[0055] Optionally, in an eighth implementation manner of the second aspect of the present invention, after establishing a coordinate system based on the first image and acquiring coordinate values ​​of each die in the wafer to be screened according to the coordinate system, the method further includes:

[0056] A division module is used to divide the wafer to be screened into a central area and an outer edge area from the origin to the outside based on the coordinate system and the first image.

[0057] Optionally, in a ninth implementation of the second aspect of the present invention, after dividing the wafer to be screened into a central area and an outer edge area from the origin to the outside in sequence based on the coordinate system and the first image, it further includes:

[0058] A judgment module, used for judging the region where each of the bare dies in the wafer to be screened is located based on each of the coordinate values;

[0059] A first marking module, configured to mark a plurality of adjacent dies of the die if the die is located in the central area and the die has been marked;

[0060] The second marking module is used to mark the die if the die is located in the outer edge area.

[0061] Optionally, in a tenth implementation manner of the second aspect of the present invention, after marking the die if the die is located in the outer edge area, the method further includes:

[0062] A third acquisition module, used for acquiring a third image of the wafer to be screened, wherein the third image is a circuit image of the marked bare die;

[0063] a second recognition module, configured to perform image recognition on the third image and determine whether the third image is a qualified circuit image;

[0064] a first placement module, configured to place the bare die into a second area if the third image is a qualified circuit image;

[0065] The second placement module is used to place the bare die in a third area if the third image is an unqualified circuit image.

[0066] A third aspect of the present invention provides an electronic device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the electronic device executes the above-mentioned screening method.

[0067] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned screening method.

[0068] In the technical solution provided by the present invention, by acquiring the distribution image of the die in the wafer to be screened, and taking the center of the wafer as the origin of the coordinate system, a two-dimensional coordinate system is established along the current horizontal and vertical directions, so as to obtain the coordinate values ​​of each die located in the wafer, and according to the coordinate values, the shooting module is controlled to shoot the corresponding die image, and the die image is input into a preset artificial intelligence model for image recognition, so as to obtain whether the die image is a die image with a complete appearance or a die image with an incomplete appearance. When the die image is a die image with a complete appearance, the clamping module is controlled to clamp the die and place the die in a good product tray, and intelligent screening is performed throughout the process to improve the efficiency of die screening; when the die image is a die image with an incomplete appearance, the die is marked and left at a distance, so that qualified products can be further screened out according to the die circuit image later, thereby improving the utilization rate of defective products. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] One or more embodiments are exemplarily described by corresponding drawings, which do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and the figures in the drawings do not constitute proportional limitations unless otherwise stated.

[0070] Figure 1 It is a schematic diagram of a first embodiment of a method for screening bare chips according to an embodiment of the present invention;

[0071] Figure 2 Schematic diagram of die distribution in wafers to be screened according to an embodiment of the present invention;

[0072] Figure 3 A schematic diagram of die distribution in which a coordinate system has been established and regions have been divided in an embodiment of the present invention;

[0073] Figure 4 is a schematic diagram of a second embodiment of a method for screening bare chips according to an embodiment of the present invention;

[0074] Figure 5 Schematic diagram of distribution of bare dies with incomplete shapes in a wafer in an embodiment of the present invention;

[0075] Figure 6 Schematic diagram of the distribution of adjacent dies of the marked dies in the wafer according to an embodiment of the present invention;

[0076] Figure 7 It is a schematic diagram of a first embodiment of a bare chip screening device according to an embodiment of the present invention;

[0077] Figure 8 It is a schematic diagram of a second embodiment of a bare die screening device according to an embodiment of the present invention;

[0078] Fig. 9 FIG. 1 is a schematic diagram of an electronic device of a bare chip according to an embodiment of the present invention. DETAILED DESCRIPTION

[0079] The main purpose of the present invention is to provide a screening method, device, equipment and storage medium, aiming to solve the technical problem that manual screening of bare chips consumes a lot of manpower and time and has low efficiency.

[0080] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0081] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1-Figure 3 , an embodiment of the method for screening bare chips in the embodiment of the present invention includes:

[0082] 101. Acquire a first image of a wafer to be screened, wherein the first image is a die distribution image 10 of the wafer to be screened;

[0083] In this embodiment, since the sizes of wafers produced in the same batch are the same, before batch screening, a bare die distribution image 10 in the wafers to be screened is first captured by a shooting device, and the bare die distribution image 10 is uploaded to a host computer. The host computer retrieves a preset image library according to the bare die distribution image 10, and matches a standard image with the same bare die shape as that in the bare die image. In the subsequent image recognition process, the standard image is used as the recognition standard, wherein several layers of circuits are completed on the wafer through epitaxy-oxidation-CVD-sputtering-photolithography-etching-ion implantation and other operations, and a plurality of bare dies are formed by cutting the wafer. The wafers to be screened in this embodiment are those after the circuits have been completed and cut, and the wafers still maintain the distribution before being cut. Therefore, the first image is the wafer after the circuits have been completed and cut to form a plurality of bare dies, and the plurality of bare die distribution images of the original wafer before cutting are retained.

[0084] 102. Establish a coordinate system based on the first image, and obtain coordinate values ​​of each die in the wafer to be screened according to the coordinate system;

[0085] In this embodiment, the host computer uses the center of the wafer as the origin of the coordinate system when acquiring the die distribution image 10, and establishes a two-dimensional coordinate system along the current horizontal and vertical directions, thereby obtaining the corresponding coordinate values ​​of each die in the wafer.

[0086] 103. Based on each of the coordinate values, obtain a second image of the wafer to be screened, wherein the second image is an image of a bare die in the wafer to be screened;

[0087] In this embodiment, the host computer controls the shooting device to refocus and adjust the focal position to shoot the image of the corresponding bare chip according to the coordinate value, so as to obtain a clearer image and improve the accuracy of image recognition. The shooting device uploads the shot bare chip image to the host computer.

[0088] 104. Identify the second image and determine whether the second image is a bare die image with a complete appearance, and obtain a determination result;

[0089] In this embodiment, the host computer performs image recognition by inputting the die image into a preset artificial intelligence model, wherein the image recognition specifically inputs the die image into a convolutional neural network and performs preprocessing such as de-meaning, normalization, and PCA / SVD dimension reduction operations in the input layer, and summarizes the original features of the image into similar fields to ensure that different features in the original features have the same dimension, and the input layer outputs an image vector; assuming that the input layer outputs a 38×38 image, and performs a convolution operation and a bias operation with a size of 5×5 pixels and a step size of 1 pixel in the convolution layer, and then activates the function, 32 feature maps of 34×3 pixels can be obtained; The 32 34×3 pixel feature maps output by the first convolutional layer are input to the first pooling layer, and the input feature maps are subjected to the maximum pooling operation. The size of the pooling block is 2×2 pixels and the step size is 1 pixel, and 32 17×17 pixel feature maps are obtained. The 32 17×17 pixel feature maps output by the pooling layer are input to the second convolutional layer, and the input feature maps are subjected to convolution and bias operations with a size of 5×5 pixels and a step size of 1 pixel, and then the activation function is used to obtain 64 13×13 pixel feature maps. The 32 17×17 pixel feature maps output by the pooling layer are input to the second convolutional layer, and the input feature maps are subjected to convolution and bias operations with a size of 5×5 pixels and a step size of 1 pixel, and then the activation function is used to obtain 64 13×13 pixel feature maps. The convolution and bias operations with a step size of 1 pixel are performed, and then the activation function is performed to obtain 64 feature maps of 13×13 pixels; the 64 feature maps of 13×13 pixels output by the second convolution layer are input into the second pooling layer, and the input feature maps are subjected to the maximum pooling operation, the size of the pooling block is 2×2 pixels, and the step size is 1 pixel, to obtain 64 feature maps with a resolution of 7×7 pixels; the activation function is used to activate each pixel point in the 64 feature maps with a resolution of 7×7 pixels output by the second pooling layer of the first fully connected layer, and the pixel value of the activated feature map is obtained, and the activated feature map is arranged in the order of columns. The 1-dimensional vector of the layer is obtained. ×3136-dimensional feature vector; then the obtained 1×3136-dimensional feature vector is input into the second fully connected layer to form a neural network, and a 1×500-dimensional feature vector is output; in the output layer, the 1×500-dimensional feature vector is calculated according to the preset calculation rules, and the calculation result is input into the Softmax classifier to obtain the classification result and output it, so that the bare die image is obtained as a bare die image with a complete appearance or an bare die image with an incomplete appearance, and intelligent image recognition is performed to improve the accuracy and effectiveness of recognition, avoid the inefficiency of artificial use of eyes and experience to identify the bare die appearance and avoid the waste of a large number of bare chips due to subjective judgment errors.

[0090] 105. Determine whether the die corresponding to the second image is complete according to the judgment result, and perform a screening operation on the die;

[0091] In this embodiment, when the die image is a complete die image, the host computer controls the clamping module to clamp the die and move the die to the good product tray, so as to realize the intelligent screening of the die, improve the efficiency and reduce the production cost of the enterprise, wherein the clamping module can be specifically a manipulator, a clamper and a grabbing clamp, etc., to grab the die. When the die image is an incomplete die image, the host computer controls the clamping module to mark the die, and then leaves the marked die in its original place, so as to further screen out qualified products according to the die circuit image in the future, so as to improve the utilization rate of defective products.

[0092] In an embodiment of the present invention, by acquiring a die distribution image 10 in a wafer to be screened, and taking the center of the wafer as the origin of the coordinate system, a two-dimensional coordinate system is established along the current horizontal and vertical directions, thereby obtaining coordinate values ​​of each die located in the wafer. According to the coordinate values, a shooting module is controlled to shoot a corresponding die image, and the die image is input into a preset artificial intelligence model for image recognition, and the die image is obtained as a die image with a complete appearance or an incomplete die image. When the die image is a die image with a complete appearance, the clamping module is controlled to clamp the die and place the die in a good product tray, and intelligent screening is performed throughout the process to improve the efficiency of die screening; when the die image is an incomplete die image, the die is marked and left at a distance, so that qualified products can be further screened out according to the die circuit image, thereby improving the utilization rate of defective products.

[0093] See also Figure 4-Figure 6 The second embodiment of the method for screening bare chips in the embodiment of the present invention includes:

[0094] 201. Acquire a first image of a wafer to be screened, wherein the first image is a die distribution image 10 of the wafer to be screened;

[0095] In this embodiment, the description of the above step 201 refers to the first embodiment, and will not be repeated in this embodiment.

[0096] 202. Taking the center of the wafer to be screened as the origin of the coordinate system, a two-dimensional coordinate system is established along the horizontal and vertical directions of the plane of the wafer to be screened, and the coordinate value of each bare die in the wafer to be screened is obtained according to the coordinate system;

[0097] In this embodiment, because the sizes of wafers in the same batch are the same, a two-dimensional coordinate system can be established along the current horizontal and vertical directions by taking the center of the circle of the wafer to be screened as the origin of the coordinate system. This is used to locate the wafers in the same batch, and then locate the position of each bare die in the wafer, so that the shooting device can accurately capture the corresponding bare die image according to the coordinate value.

[0098] 203. Based on each of the coordinate values, control a photographing device to photograph an image of a die corresponding to the coordinate value in the wafer to be screened to obtain the die image;

[0099] In this embodiment, the shooting device can be a fully automatic center-focus camera or a high-definition camera to improve the clarity of the captured image, which is conducive to more accurate image recognition by the preset artificial intelligence model; the host computer controls the shooting device to refocus and adjust the focal position to shoot the image of the corresponding bare chip according to the coordinate value, so as to obtain a clearer image and improve the accuracy of image recognition. The shooting device uploads the captured bare chip image to the host computer.

[0100] Optionally, in one embodiment, after step 203, the method further includes: dividing the wafer to be screened into a central area 12 and an outer edge area 11 from the origin to the outside in sequence based on the coordinate system and the first image;

[0101] Based on the coordinate values, determining the region where each of the bare dies in the wafer to be screened is located;

[0102] If the die is located in the central area 12 and the die has been marked, then a plurality of adjacent dies 16 of the die are marked;

[0103] If the die is located in the outer edge area 11 , the die is marked.

[0104] In this embodiment, the host computer divides the die distribution image 10 into a central area 12 and an outer area 11 from the origin of the coordinate system to the outside according to the coordinate system. At the same time, the boundaries of the two areas are determined by taking a fixed ratio of the maximum number of rows and columns of the wafer to be screened. Of course, it can also be directly demarcated artificially, wherein the size of the specific divided area depends on the size of the wafer and the size of the die. According to the corresponding coordinate values ​​of each die, the area where each die in the wafer to be screened is located is determined. When the die is located in the central area 12 and the die has been marked, multiple adjacent die 16 of the die are marked; when the die is located in the outer area 11, the die is marked. Marking the die is conducive to improving the efficiency of subsequent screening of the die. At the same time, because the central area 12 includes most of the complete die, the die located in the central area 12 is screened first, and after the die located in the central area 12 is screened, the die located in the outer area 11 is screened. Filter by area to improve screening efficiency.

[0105] Optionally, in one embodiment, after step 203, after “if the die is located in the outer edge area 11, mark the die”, the following further comprises:

[0106] Acquire a third image of the wafer to be screened, wherein the third image is a circuit image of the marked bare die 15;

[0107] Performing image recognition on the third image, and determining whether the third image is a qualified circuit image;

[0108] If the third image is a qualified circuit image, placing the bare die in the second area;

[0109] If the third image is an unqualified circuit image, the bare die is placed in a third area.

[0110] Specifically, because the incomplete appearance of the die in the central area 12 is usually caused by physical damage, it will affect other die around the incomplete die. Therefore, after the host computer recognizes the die appearance image, it marks the die with incomplete appearance, obtains the adjacent die 16 of the incomplete die in the coordinate system, and further screens out qualified die by recognizing the circuit image of the adjacent die 16. Before the shooting device captures the die image, the lighting module is used to illuminate the die to be photographed, so that the image information on the die surface can be more clearly captured by the shooting device, and the light source can be used to reflect not only the specific shape of the die, but also the circuit on the die, to capture a clear appearance image and circuit image, wherein the lighting module includes a concave-convex lens with the focal length of the light beam or a plane mirror that changes the direction of the light path.

[0111] In this embodiment, the third image is a circuit image of the marked die 15, the second area is a substandard tray, and the third area is a defective tray; the host computer respectively obtains the coordinate values ​​of multiple die adjacent to the marked die 15, and controls the shooting device to shoot the circuit image of each adjacent die 16 according to the obtained coordinate values, so as to obtain multiple circuit images of adjacent die 16; the circuit images of the multiple marked die 15 are input into the preset artificial intelligence model for image recognition, and the third image is obtained as a qualified circuit image or an unqualified circuit image. When it is a qualified circuit image, the host computer controls the clamping module to clamp the adjacent die 16 and place it in the substandard tray; it is beneficial for the enterprise to apply the die in the substandard tray to the production that meets the requirements, and improve the utilization rate of defective die. When it is an unqualified circuit image, the host computer controls the clamping module to clamp the adjacent die 16 and place it in the defective tray to complete the screening.

[0112] Optionally, in another embodiment, the method further includes: dividing the first image into a plurality of the second images according to a preset segmentation rule;

[0113] Based on each of the coordinate values, acquiring a second image corresponding to the coordinate value;

[0114] In this embodiment, the host computer divides the die distribution map in the wafer to be screened into individual die images according to a preset segmentation rule, and then extracts the die images corresponding to the coordinate values ​​according to the coordinate values ​​of the plurality of dies previously acquired. The preset segmentation rule may be to directly set the image size of the required second image to segment the first image, or to perform cutting according to the boundaries between adjacent dies in the first image.

[0115] 204. Identify the second image and determine whether the second image is a bare die image with a complete appearance, and obtain a determination result.

[0116] In this embodiment, the description of the above step 204 refers to the first embodiment, and will not be repeated in this embodiment.

[0117] Optionally, in one embodiment, before step 204, the following steps are further included:

[0118] Inputting the second image into a preset artificial intelligence model;

[0119] In this embodiment, the method of establishing the preset artificial intelligence model includes but is not limited to, training a large number of images in a bare die image library with information tags (including bare die images with complete appearance and bare die images with incomplete appearance) through a CNN convolutional neural network model to establish an image model; the upper computer inputs the bare die image into the preset artificial intelligence model for image recognition, determines whether the appearance of the bare die image is a complete image, and performs intelligent image recognition to improve the accuracy and effectiveness of recognition, thereby avoiding the inefficiency of bare die appearance recognition based on human eyes and experience and the waste of a large number of bare chips due to subjective judgment errors.

[0120] 205. If the second image is an image of a die with a complete appearance, determine that the die corresponding to the second image is complete, and place the die in the first area;

[0121] In this embodiment, the first area is a good product tray, and the second image is a bare die image. When the bare die image is a bare die image with a complete appearance, the host computer controls the clamping module to clamp the bare die and move the bare die to the good product tray, thereby realizing intelligent screening of bare die, improving efficiency, and reducing enterprise production costs. Among them, the clamping module can be specifically a manipulator, a clamper, a grabbing clamp, etc., to grab the bare die.

[0122] 206. If the second image is an image of a die 13 with an incomplete appearance, determine that the die corresponding to the second image is incomplete, and mark the die;

[0123] In this embodiment, when the die image is an incomplete die image, the host computer controls the clamping module to mark the die, and then leaves the marked die in place, so that qualified products can be further screened out according to the die circuit image, thereby improving the utilization rate of defective products.

[0124] In this embodiment, when the circuit of a conventional type of bare chip is damaged, it will be scrapped because it cannot realize normal functions, but the bare chip of flash memory type (NandFlash, NorFlash, etc.) is different. There can be multiple planes (Plane) in the flash memory bare chip, and there will be multiple blocks (Block) in the plane. The block is composed of multiple pages, and its bottom layer is a data storage unit formed by the intersection of the bit line WL (WordLine) and the word line BL (BitLine). The entire bare chip can be a combination of multiple storage units. Therefore, when the circuit of the bare chip of the flash memory type is damaged, it may be that only part of the bit line and word line are damaged, that is, part of the storage unit is damaged, and the remaining storage units are still in good condition, so the bare chip 13 with incomplete appearance has the possibility of recycling. Therefore, after the bare chips in the central area and the outer edge area are screened once, the bare chips 13 with incomplete appearance are screened by identifying the circuit image of the bare chip, and the qualified bare chips are further selected, and the clamping module is controlled to clamp the bare chip into the sub-good product tray. Improve the utilization rate of defective bare chips.

[0125] In this embodiment, a camera is used to capture the image of the die and perform image recognition on the die image, so that the die screening is made intelligent. A coordinate system is established on the die distribution image 10 obtained in the wafer to be screened, so as to facilitate the subsequent positioning and area division of the die. First, the central area 12 containing most of the die with complete appearance is screened, and then the outer edge area 11 with most of the die with incomplete appearance is screened to improve the screening efficiency. Then, the marked die 15 after screening is screened to screen out the die with the circuit meeting the requirements, so as to improve the utilization rate of defective die.

[0126] The above describes the bare chip screening method in the embodiment of the present invention. The following describes the bare chip screening device in the embodiment of the present invention. Figure 7 The first embodiment of the device for screening bare chips in the embodiment of the present invention comprises:

[0127] A first acquisition module 301 is used to acquire a first image of the wafer to be screened, wherein the first image is a die distribution image 10 of the wafer to be screened;

[0128] An establishing module 302 is used to establish a coordinate system based on the first image, and obtain the coordinate value of each die in the wafer to be screened according to the coordinate system;

[0129] A second acquisition module 303 is used to acquire a second image of the wafer to be screened based on each of the coordinate values, wherein the second image is an image of a bare die in the wafer to be screened;

[0130] A first recognition module 304 is used to perform image recognition on the second image, and determine whether the second image is a bare die image with a complete appearance, and obtain a determination result;

[0131] The screening module 305 is used to determine whether the die corresponding to the second image is complete according to the judgment result, and perform a screening operation on the die.

[0132] In this embodiment, by acquiring the die distribution image 10 in the wafer to be screened, and taking the center of the wafer as the origin of the coordinate system, a two-dimensional coordinate system is established along the current horizontal and vertical directions, thereby obtaining the coordinate values ​​of each die located in the wafer. According to the coordinate values, the shooting module is controlled to shoot the corresponding die image, and the die image is input into the preset artificial intelligence model for image recognition, and the die image is obtained as a die image with a complete appearance or an incomplete appearance. When the die image is a die image with a complete appearance, the clamping module is controlled to clamp the die and place the die in a good product tray, and intelligent screening is performed throughout the process to improve the efficiency of die screening; when the die image is an incomplete die image, the die is marked and left at a distance, so that qualified products can be further screened out according to the die circuit image later, thereby improving the utilization rate of defective products.

[0133] See also Figure 8 The second embodiment of the bare chip screening device in the embodiment of the present invention includes:

[0134] A first acquisition module 301 is used to acquire a first image of the wafer to be screened, wherein the first image is a die distribution image 10 of the wafer to be screened;

[0135] An establishing module 302 is used to establish a coordinate system based on the first image, and obtain the coordinate value of each die in the wafer to be screened according to the coordinate system;

[0136] A second acquisition module 303 is used to acquire a second image of the wafer to be screened based on each of the coordinate values, wherein the second image is an image of a bare die in the wafer to be screened;

[0137] A first recognition module 304 is used to perform image recognition on the second image, and determine whether the second image is a bare die image with a complete appearance, and obtain a determination result;

[0138] A screening module 305 is used to determine whether the die corresponding to the second image is complete according to the judgment result, and perform a screening operation on the die;

[0139] An illumination module 306, used to illuminate the die to be photographed;

[0140] An input module 307, used to input the second image into a preset artificial intelligence model;

[0141] A division module 308 is used to divide the wafer to be screened into a central area 12 and an outer edge area 11 from the origin to the outside based on the coordinate system and the first image;

[0142] A judgment module 309 is used to judge the region where each of the dies in the wafer to be screened is located based on each of the coordinate values;

[0143] A first marking module 310 is configured to mark a plurality of adjacent dies 16 of the die if the die is located in the central area 12 and the die has been marked;

[0144] A second marking module 311 is used to mark the die if the die is located in the outer edge area 11;

[0145] A third acquisition module 312 is used to acquire a third image of the wafer to be screened, wherein the third image is a circuit image of the marked bare die 15;

[0146] The second recognition module 313 is used to perform image recognition on the third image and determine whether the third image is a qualified circuit image;

[0147] A first placement module 314, configured to place the die in a second area if the third image is a qualified circuit image;

[0148] The second placement module 315 is configured to place the die in a third area if the third image is an unqualified circuit image.

[0149] Optionally, in one embodiment, the second obtaining module 303 includes:

[0150] a segmentation unit, configured to segment the first image into a plurality of the second images according to a preset segmentation rule;

[0151] The first acquisition unit is used to acquire a second image corresponding to the coordinate value based on each of the coordinate values.

[0152] Optionally, in one embodiment, the screening module 305 includes:

[0153] a first determining unit, configured to determine that the die corresponding to the second image is complete if the second image is an image of a die with a complete appearance, and place the die in a first area;

[0154] The marking unit is configured to determine that the die corresponding to the second image is incomplete and mark the die if the second image is an image of the die 13 with an incomplete appearance.

[0155] Optionally, in one embodiment, the establishing module 302 includes:

[0156] The establishing unit is used to establish a two-dimensional coordinate system along the horizontal direction and the vertical direction of the plane of the wafer to be screened by taking the center of the circle of the wafer to be screened as the origin of the coordinate system.

[0157] The judgment result includes: the second image is an image of a die with a complete appearance; or the second image is an image of a die 13 with an incomplete appearance.

[0158] In this embodiment, before the shooting device shoots the image of the die, the lighting module is used to illuminate the die to be photographed, so that the image information of the die surface can be more clearly acquired by the shooting device. Not only can the light source be used to reflect the specific shape of the die, but also the circuit on the die can be used to reflect the circuit, so as to capture a clear appearance image and circuit image, and perform image recognition on the die image, so as to make the die screening intelligent; and by establishing a coordinate system on the die distribution image 10 obtained in the wafer to be screened, so as to facilitate the subsequent positioning and area division of the die, firstly, after the die screening is completed for the central area 12 containing most of the die with complete appearance, the outer edge area 11 with most of the die with incomplete appearance is screened to improve the screening efficiency; and then the marked die 15 after screening is screened to screen out the die with the circuit meeting the requirements, so as to improve the utilization rate of defective die.

[0159] above Figure 7 and Figure 8 The bare chip screening device in the embodiment of the present invention is described in detail from the perspective of modular functional entities, and the bare chip electronic device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0160] Fig. 91 is a schematic diagram of the structure of a bare chip electronic device provided by an embodiment of the present invention. The electronic device 500 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 510 (for example, one or more processors) and a memory 520, and one or more storage media 530 (for example, one or more mass storage devices) storing application programs 533 or data 532. Among them, the memory 520 and the storage medium 530 can be short-term storage or permanent storage. The program stored in the storage medium 530 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the electronic device 500. Furthermore, the processor 510 can be configured to communicate with the storage medium 530 to execute a series of instruction operations in the storage medium 530 on the electronic device 500.

[0161] The electronic device 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input and output interfaces 560, and / or one or more operating systems 531, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Fig. 9 The structure of the electronic device shown does not constitute a limitation of the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0162] The present invention also provides a bare chip electronic device, which includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor executes the steps of the screening method in the above embodiments.

[0163] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium, and when the instructions are executed on a computer, the computer executes the steps of the screening method.

[0164] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0165] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0166] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for screening bare chips, characterized in that: The screening method comprises: Acquire a first image of a wafer to be screened, wherein the first image is a distribution image of bare dies in the wafer to be screened, and the wafer to be screened is a storage wafer; Based on the first image, a coordinate system is established with the center of the wafer to be screened as the origin of the coordinate system, and the coordinate value of each bare die in the wafer to be screened is obtained according to the coordinate system; Based on each of the coordinate values, obtaining a second image of the wafer to be screened, wherein the second image is an image of a bare die in the wafer to be screened; Performing image recognition on the second image, and determining whether the second image is a bare die image with a complete appearance, to obtain a determination result; Determine whether the die corresponding to the second image is complete according to the judgment result, and perform a screening operation on the die; If the second image is an image of a die with an incomplete appearance, determining that the die corresponding to the second image is incomplete, and marking the die; Based on the coordinate system and the first image, the wafer to be screened is divided into a central area and an outer edge area from the origin to the outside; Determine the region where each of the dies in the wafer to be screened is located, and first screen the dies located in the central region; If the die is located in the central area and the die has been marked, marking a plurality of adjacent dies of the die; If the die is located in the outer edge area, marking the die; Acquire a third image of the wafer to be screened, wherein the third image is a circuit image of the marked bare die; Perform image recognition on the third image, and determine whether the third image is a qualified circuit image.

2. The screening method according to claim 1, characterized in that The step of acquiring a second image of the wafer to be screened based on each of the coordinate values, wherein the second image is an image of a bare die in the wafer to be screened, comprises: Based on each of the coordinate values, a photographing device is controlled to photograph an image of a die in the wafer to be screened that corresponds to the coordinate value, so as to obtain the die image.

3. The screening method according to claim 1, characterized in that The step of acquiring a second image of the wafer to be screened based on each of the coordinate values, wherein the second image is an image of a bare die in the wafer to be screened, comprises: According to a preset segmentation rule, segmenting the first image into a plurality of the second images; Based on each of the coordinate values, a second image corresponding to the coordinate value is acquired.

4. The screening method according to claim 1, characterized in that Before performing image recognition on the second image and determining whether the second image is a bare die image with a complete appearance and obtaining a determination result, the method further includes: The second image is input into a preset artificial intelligence model.

5. The screening method according to claim 1, characterized in that The judgment results include: The second image is an image of a die with a complete appearance; or, the second image is an image of a die with an incomplete appearance.

6. The screening method according to claim 5, characterized in that The step of determining whether the die corresponding to the second image is complete according to the judgment result and performing a screening operation on the die includes: If the second image is an image of a die with a complete appearance, it is determined that the die corresponding to the second image is complete, and the die is placed in the first area.

7. The screening method according to claim 6, characterized in that The step of establishing a coordinate system includes: establishing a two-dimensional coordinate system along the horizontal direction and the vertical direction of the plane of the wafer to be screened.

8. The screening method according to claim 1, characterized in that After marking the die if the die is located in the outer edge area, the method further includes: If the third image is a qualified circuit image, placing the bare die in the second area; If the third image is an unqualified circuit image, the bare die is placed in a third area.

9. A bare chip screening device, characterized in that: The screening device comprises: A first acquisition module is used to acquire a first image of the wafer to be screened, wherein the first image is a distribution image of bare dies in the wafer to be screened; An establishment module is used to establish a coordinate system based on the first image and taking the center of the wafer to be screened as the origin of the coordinate system, and obtain the coordinate value of each bare die in the wafer to be screened according to the coordinate system; A second acquisition module, configured to acquire a second image of the wafer to be screened based on each of the coordinate values, wherein the second image is an image of a bare die in the wafer to be screened; a recognition module, configured to perform image recognition on the second image, and determine whether the second image is a bare die image with a complete appearance, and obtain a determination result; a screening module, configured to determine whether the die corresponding to the second image is complete according to the judgment result, and perform a screening operation on the die; A division module, used for dividing the wafer to be screened into a central area and an outer edge area from the origin to the outside in sequence based on the coordinate system and the first image; A judgment module, for judging the region where each of the dies in the wafer to be screened is located, and firstly screening the dies located in the central region; A first marking module, configured to mark a plurality of adjacent dies of the die if the die is located in the central area and the die has been marked; A second marking module, configured to mark the die if the die is located in the outer edge area; A third acquisition module, used to acquire a third image of the wafer to be screened, wherein the third image is a circuit image of the marked bare die; The second recognition module is used to perform image recognition on the third image and determine whether the third image is a qualified circuit image.

10. The screening device according to claim 9, characterized in that: The screening device also includes: An illumination module is used to illuminate the die to be photographed.

11. An electronic device, characterized in that: The electronic device comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory to enable the electronic device to perform the screening method according to any one of claims 1 to 8.

12. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by a processor, the screening method according to any one of claims 1 to 8 is implemented.

Citation Information

Cited By

  • Chip identification and positioning algorithm based on coordinate system conversion

    CN119399204A

  • A chip identification and positioning algorithm based on coordinate system transformation

    CN119399204B