Semiconductor device detection method, device, electronic device and storage medium

Abstract

This application provides a semiconductor device inspection method, apparatus, electronic device, and storage medium. The method comprises: acquiring two perpendicular DR images of a semiconductor device to be inspected; obtaining a three-dimensional volume of the semiconductor device to be inspected corresponding to the two perpendicular DR images based on a preset three-dimensional volume reconstruction model; and performing defect detection on the semiconductor device to be inspected based on the three-dimensional volume. The preset three-dimensional volume reconstruction model obtains the three-dimensional volume by fusing large-scale structure recovery and small-scale detail reconstruction. This method enables efficient and high-precision defect detection of semiconductor devices while saving manpower and material resources.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of detection technology, and in particular to a semiconductor device detection method, apparatus, electronic device, and storage medium. Background Art

[0002] In recent years, with the rapid development of the semiconductor industry, the design and manufacturing processes of semiconductor devices have become increasingly sophisticated and complex to meet the needs of today. Therefore, early detection of internal defects has become particularly important. This not only reduces unnecessary losses but also improves efficiency, thereby gaining a competitive advantage in time.

[0003] Traditional nondestructive testing methods rely primarily on manual inspection, which is costly and inefficient. Therefore, there is an urgent need for an efficient, intelligent nondestructive testing method suitable for mass production. Although DR technology is widely used in medical examinations, with the advantages of real-time performance and low cost, and can detect device defects to a certain extent, small semiconductor devices with complex, dense, and obscured internal structures still require the use of computed tomography technology to reconstruct the device's three-dimensional structure for defect detection. However, the principle of computed tomography technology is based on circumferential projection data, which cannot meet the needs of real-time online reconstruction.

[0004] Therefore, there is an urgent need for a semiconductor device inspection method that can improve defect detection speed and accuracy while saving manpower and material resources. Summary of the Invention

[0005] In view of this, the present application provides a semiconductor device detection method, apparatus, electronic equipment and storage medium, which can realize defect detection of semiconductor devices with high efficiency and high precision while saving manpower and material resources.

[0006] To solve the above technical problems, the technical solution of this application is implemented as follows:

[0007] In one embodiment, a semiconductor device detection method is provided, the method comprising:

[0008] Acquire two DR images with a vertical relationship for the semiconductor device to be inspected;

[0009] Obtaining a three-dimensional volume of the semiconductor device to be inspected corresponding to the two DR images having a perpendicular relationship based on a preset three-dimensional volume reconstruction model;

[0010] performing defect detection on the semiconductor device to be inspected based on the three-dimensional volume;

[0011] The preset 3D volume reconstruction model includes: a channel number enhancement module, a structure recovery module, a detail enhancement module and a fusion module;

[0012] The channel number increasing module is used to increase the channel number of the two DR images having a vertical relationship while ensuring that the spatial resolution remains unchanged; and to fuse the two DR images after the channel number increase to obtain a first fusion result;

[0013] The structure recovery module is used to perform large-scale structure recovery on the first fusion result to obtain first feature information;

[0014] The detail enhancement module is used to reconstruct small-scale details of the first fusion result to obtain second feature information;

[0015] The fusion module is used to fuse the first feature information and the second feature information to obtain a second fusion result; the second fusion result is the three-dimensional volume of the semiconductor device to be inspected corresponding to the two DR images with a vertical relationship.

[0016] The generation of the preset three-dimensional volume reconstruction model includes:

[0017] Generate a 3D volume reconstruction model to be trained;

[0018] Acquire a training dataset; wherein the training dataset includes multiple data pairs, each data pair including two DR images with a perpendicular relationship and a reference three-dimensional volume; the DR image data with a perpendicular relationship are acquired from an acquisition dataset consisting of axially acquired DR images of a reference semiconductor device, and the reference three-dimensional volume is obtained by performing FDK reconstruction based on the acquisition dataset;

[0019] The three-dimensional volume reconstruction model to be trained is trained based on the training data set to obtain a preset three-dimensional volume reconstruction model.

[0020] Wherein, when training the three-dimensional volume reconstruction model to be trained based on the training data set, a preset loss function is introduced;

[0021] When N consecutive values of the preset loss function are all less than a preset loss threshold, the training of the to-be-trained three-dimensional volume reconstruction model is terminated to obtain a preset three-dimensional volume reconstruction model;

[0022] Wherein, N is an integer greater than 1; the preset loss function is a weighted sum of a squared error loss function and a structural exponential loss function; the value of the preset loss function is calculated based on the reference three-dimensional volume and the three-dimensional volume obtained by the three-dimensional volume reconstruction model to be trained using the preset loss function.

[0023] in,

[0024] The structure recovery module uses an encoder to upsample the first fusion result to obtain a potentially compact representation feature; uses a decoder to receive the feature information output by the encoder, reconstructs the potential information through an upsampling operation, and obtains the first feature information; and introduces jump connections at different scales during the encoding and decoding process.

[0025] in,

[0026] The detail enhancement module obtains feature maps with different receptive fields through three parallel dilated convolutions with different dilation rates on the first fusion result, connects the feature maps with different receptive fields, and uses a convolution layer to restore the number of channels corresponding to the connection result to obtain a first processing result; obtains a second processing result through two convolution layers on the first fusion result; and fuses the first processing result and the second processing result to obtain second feature information.

[0027] in,

[0028] The detail enhancement module fusing the first processing result and the second processing result to obtain second feature information includes:

[0029] Performing global pooling on the first processing result to obtain a global feature vector, and generating a first attention weight corresponding to the first processing result using a fully connected layer and a softmax function;

[0030] Performing global pooling on the second processing result to obtain a global feature vector, and generating a second attention weight corresponding to the second processing result using a fully connected layer and a softmax function;

[0031] Second feature information is determined based on the first processing result, the first attention weight, the second processing result and the second attention weight.

[0032] In another embodiment, a semiconductor device detection apparatus is provided, the apparatus comprising:

[0033] An acquisition unit, configured to acquire two DR images having a vertical relationship with respect to the semiconductor device to be inspected;

[0034] An acquisition unit is configured to obtain, based on a preset three-dimensional volume reconstruction model, the three-dimensional volume of the semiconductor device to be inspected corresponding to the two DR images with a vertical relationship; wherein the preset three-dimensional volume reconstruction model includes: a channel number improvement module, a structure recovery module, a detail enhancement module, and a fusion module; the channel number improvement module is configured to respectively improve the channel number of the two DR images with a vertical relationship while ensuring that the spatial resolution remains unchanged; and to fuse the two DR images after the channel number improvement to obtain a first fusion result; the structure recovery module is configured to perform large-scale structure recovery on the first fusion result to obtain first feature information; the detail enhancement module is configured to perform small-scale detail reconstruction on the first fusion result to obtain second feature information; the fusion module is configured to fuse the first feature information and the second feature information to obtain a second fusion result; the second fusion result is the three-dimensional volume of the semiconductor device to be inspected corresponding to the two DR images with a vertical relationship;

[0035] a detection unit, configured to perform defect detection on the semiconductor device to be detected based on the three-dimensional volume;

[0036] Wherein, the device further comprises:

[0037] A training unit is configured to generate a three-dimensional volume reconstruction model to be trained; obtain a training data set; wherein the training data set includes multiple data pairs, each data pair including two DR images with a vertical relationship and a three-dimensional volume; the DR image data with a vertical relationship are obtained from a data set consisting of DR images of a reference semiconductor device acquired axially, and the three-dimensional volume is obtained by FDK reconstruction based on the data set; the three-dimensional volume reconstruction model to be trained is trained based on the training data set to obtain a preset three-dimensional volume reconstruction model.

[0038] In another embodiment, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements a semiconductor device detection method when executing the program.

[0039] In another embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, a semiconductor device detection method is implemented.

[0040] As can be seen from the above technical solution, the above embodiment provides a method for acquiring two perpendicular DR images of a semiconductor device to be inspected, and then using a preset 3D volume reconstruction model to obtain the 3D volume of the semiconductor device to be inspected, thereby enabling defect detection of the semiconductor device. The preset 3D volume reconstruction model is capable of recovering large-scale information and reconstructing small-scale information, fusing the two results to obtain a high-precision 3D volume. Therefore, defect detection of semiconductor devices can be achieved efficiently and accurately while saving manpower and material resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0042] Figure 1 Schematic diagram of the process of establishing a preset three-dimensional volume reconstruction model in an embodiment of the present application;

[0043] Figure 2 This is a schematic diagram of the structure of a preset three-dimensional volume reconstruction model in an embodiment of the present application;

[0044] Figure 3 This is a schematic diagram of the structure of the channel number increasing module in the embodiment of the present application;

[0045] Figure 4 This is a schematic diagram of the structure recovery module in the embodiment of the present application;

[0046] Figure 5 This is a schematic diagram of the detail enhancement module structure in an embodiment of the present application;

[0047] Figure 6 This is a schematic diagram of the semiconductor device detection process in an embodiment of the present application;

[0048] Figure 7 This is a schematic diagram of a semiconductor device detection structure in an embodiment of the present application;

[0049] Figure 8 A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0051] The terms "first", "second", "third", "fourth", etc. (if any) in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe the order or precedence of the objects. It should be understood that the numbers used in this way are interchangeable where appropriate, so that the implementation of the invention described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus 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 that are inherent to these processes, methods, products or apparatus.

[0052] The following specific embodiments are used to describe the technical solution of the present invention in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0053] Traditional nondestructive testing methods rely primarily on manual inspection, which is costly and inefficient. Therefore, there is an urgent need for an efficient, intelligent nondestructive testing method suitable for mass production. Although DR technology is widely used in medical examinations, with the advantages of real-time performance and low cost, and can detect device defects to a certain extent, small semiconductor devices with complex, dense, and obscured internal structures still require the use of computed tomography technology to reconstruct the device's three-dimensional structure for defect detection. However, the principle of computed tomography technology is based on circumferential projection data, which cannot meet the needs of real-time online reconstruction.

[0054] For X-ray CT imaging, it is usually necessary to collect 360-degree projection data of the object, and the reconstruction algorithm is complex and time-consuming, which cannot meet the real-time detection requirements. In order to improve the detection speed, CT imaging based on sparse perspective is generally used, mainly including reconstruction algorithms based on compressed sensing, reconstruction methods based on optimization algorithms, reconstruction technologies based on model assistance, and reconstruction methods based on deep learning. However, these methods have their own limitations, such as slow imaging speed, imaging effects that decrease with decreasing angles, and imaging results that rely on model construction. Among them, for the three-dimensional volume reconstruction problem based on deep learning methods, there is currently a single-projection three-dimensional reconstruction network, but the detailed structure in the reconstruction results of devices with rich internal information is not well restored. Although the dual-projection three-dimensional reconstruction network uses orthogonal projection information to complement each other, it still has the problem of not being able to restore small-scale details well, which limits the quality of the reconstructed volume.

[0055] Therefore, there is an urgent need for a semiconductor device inspection method that can improve defect detection speed and detection accuracy while saving manpower and material resources.

[0056] To address the above issues, an embodiment of the present application provides a semiconductor device inspection method. By simply acquiring two perpendicular DR images of the semiconductor device to be inspected, a preset 3D volume reconstruction model can be used to obtain the 3D volume of the semiconductor device to be inspected, thereby enabling defect detection of the semiconductor device. The preset 3D volume reconstruction model is capable of recovering large-scale information and reconstructing small-scale information, fusing the results of the two to obtain a high-precision 3D volume. Therefore, defect detection of semiconductor devices can be achieved efficiently and accurately while saving manpower and material resources.

[0057] In the embodiment of the present application, before performing semiconductor device testing, it is necessary to first establish a preset three-dimensional volume reconstruction model. The specific establishment process is as follows:

[0058] See also Figure 1 , Figure 1 This is a schematic diagram of the process of establishing a preset 3D volume reconstruction model in an embodiment of the present application. The specific steps are:

[0059] Step 101: Generate a 3D volume reconstruction model to be trained.

[0060] Step 102: Obtain a training data set.

[0061] Among them, the training data set includes multiple data pairs, each data pair includes two DR images with a vertical relationship and a reference three-dimensional volume; the DR image data with a vertical relationship are obtained from an acquisition data set consisting of DR images of reference semiconductor devices acquired axially, and the reference three-dimensional volume is obtained by FDK reconstruction based on the acquisition data set.

[0062] The training data set obtained can be a pre-stored data set or a collected data set.

[0063] The implementation for acquiring the training data set is as follows:

[0064] Circumferential DR image acquisition is performed on semiconductor devices (reference objects) with various defect types. 360 DR images are acquired from different directions for each semiconductor device.

[0065] During the acquisition process, relevant parameters such as tube voltage, tube current, exposure time, and position during detection are guaranteed to remain unchanged.

[0066] The acquired DR images can be preprocessed first, and the preprocessing includes:

[0067] The captured DR images are cropped, logarithmized, and normalized based on chip and mechanical parameters. Cropping is performed to 256×256×360 pixels. Normalization involves normalizing the image and voxel grayscale values to [0, 1]. The reference 3D volume of the reference object is then obtained, and model training is performed.

[0068] Obtain the reference 3D volume for each reference object, including:

[0069] A mature algorithm such as FDK can be used to obtain the reference 3D volume of each reference object based on 360 DR images of the reference object.

[0070] Step 103 : training the 3D volume reconstruction model to be trained based on the training data set to obtain a preset 3D volume reconstruction model.

[0071] The Adam optimizer is used during training, the learning rate starts from 1e-3, and the learning rate decay value is set to 0.8.

[0072] In the training process, the present embodiment also introduces a preset loss function; the preset loss function uses a combination of the mean square error (MSE) loss function and the structural similarity index (SSIM) loss function to quantify the difference between the three-dimensional volume obtained by the three-dimensional volume reconstruction model to be trained and the reference three-dimensional volume. The preset loss function is specifically a weighted sum of the mean square error loss function and the structural index loss function; the value of the preset loss function is calculated based on the reference three-dimensional volume and the three-dimensional volume obtained by the three-dimensional volume reconstruction model to be trained using the preset loss function. Specifically:

[0073] The squared error loss function is calculated by calculating the sum of squared errors between the reference 3D volume and the 3D volume obtained by the 3D volume reconstruction model to be trained. This loss will achieve a good numerical reconstruction of the 3D volume. The specific expression is as follows:

[0074]

[0075] Among them, loss MSE is the value of the error square sum function, y0 is the reference three-dimensional volume, is a three-dimensional volume obtained by the three-dimensional volume reconstruction model to be trained;

[0076] The value of the structural exponential loss function is determined using the structural exponential loss function based on the reference 3D volume and the 3D volume obtained by the 3D volume reconstruction model to be trained. This loss will achieve a good reconstruction of the 3D volume that is consistent with human vision. The specific expression is as follows:

[0077]

[0078] Among them, loss SSIM is the value of the structural index loss function, y0 is the reference three-dimensional volume, is a three-dimensional volume obtained by the three-dimensional volume reconstruction model to be trained;

[0079] The default loss function is the weighted sum of the squared error loss function and the structural exponential loss function, which is expressed as follows:

[0080] loss = αloss MSE +βloss SSIM

[0081] Among them, loss is the value of the preset loss function, α and β are weight values, and the specific values of α and β are not restricted and can be set according to actual needs.

[0082] When the values of N consecutive preset loss functions are all less than the preset loss threshold, the training of the to-be-trained three-dimensional volume reconstruction model is terminated to obtain a preset three-dimensional volume reconstruction model; wherein N is an integer greater than 1.

[0083] In the embodiment of the present application, the structures of the 3D volume reconstruction model to be trained and the preset 3D volume reconstruction model are the same, and a detailed description is given taking the preset 3D volume reconstruction model structure as an example.

[0084] See also Figure 2 , Figure 2 This is a schematic diagram of the structure of the preset three-dimensional volume reconstruction model in an embodiment of the present application. Figure 2 The preset 3D volume reconstruction model in the package includes: channel number enhancement module, structure recovery module, detail enhancement module and fusion module;

[0085] Among them, the channel number enhancement module is used to enhance the channel numbers of two DR images with a vertical relationship while ensuring that the spatial resolution remains unchanged; and fuse the two DR images after the channel numbers are enhanced to obtain a first fusion result.

[0086] See also Figure 3 , Figure 3 This is a schematic diagram of the channel number increasing module structure in an embodiment of the present application. Figure 3 The channel number enhancement module includes two dense residual modules and a fuser 313 , where the two dense residual modules are represented as a first dense residual module 311 and a second dense residual module 312 .

[0087] Two DR images with a vertical relationship are respectively recorded as the first DR image 301 and the second DR image 302, and the input channel number enhancement module, specifically, the first DR image 301 is input into the first dense residual module 311, and the second DR image 302 is input into the second dense residual module 312; the first dense residual module 311 and the second dense residual module 312 respectively enhance the channel number of the input DR images while keeping the spatial resolution unchanged, and the fusion unit 313 fuses the results output by the first dense residual module 311 and the second dense residual module 312 to output a first fusion result.

[0088] The structure recovery module is used to perform large-scale structure recovery on the first fusion result to obtain first feature information.

[0089] The structure recovery module uses the encoder to upsample the first fusion result to obtain potential compact representation features; uses the decoder to receive the feature information output by the encoder, reconstructs the potential information through upsampling operations, and obtains the first feature information; and introduces jump connections at different scales in the encoding and decoding process.

[0090] See also Figure 4 , Figure 4 This is a schematic diagram of the structure recovery module in an embodiment of the present application. Figure 4 The structure recovery module in [1] includes: encoder and decoder;

[0091] The decoder consists of four downsampling blocks, which are used to learn potentially compact representations. These four downsampling modules capture the key features of the input information while removing redundancy. Assuming the input corresponds to the first fusion result of 8, 128, 256, 256, the information after the first downsampling block is 8, 128, 128, 128; after the second downsampling block, it is 8, 128, 64, 64; after the third downsampling block, it is 8, 128, 32, 32; and after the fourth downsampling block, it is 8, 128, 16, 16. This is just an example, and the specific implementation is not limited to the above information. The first data 8 represents the batch size, the second data 128 represents the number of channels, and the third and fourth data represent the length and width, which are scaled down by multiples. Each downsampling block consists of two residual blocks and a maximum pooling layer.

[0092] The encoder consists of four upsampling blocks, which reconstruct the latent information through upsampling operations.

[0093] At the same time, in order to retain the original important information and restore the fine-grained spatial details lost in the encoding process, the first fusion result is obtained through the above-mentioned structure recovery module, that is, jump connections at different scales are introduced in the encoding and decoding process, such as Figure 4 As shown by the dotted line in .

[0094] The detail enhancement module is used to reconstruct small-scale details of the first fusion result to obtain the second feature information, specifically:

[0095] The detail enhancement module obtains feature maps with different receptive fields through three parallel dilated convolutions with different dilation rates on the first fusion result, connects the feature maps with different receptive fields, and uses the convolution layer to restore the number of channels corresponding to the connection result to obtain the first processing result; obtains the second processing result through two convolution layers on the first fusion result; and fuses the first processing result and the second processing result to obtain the second feature information.

[0096] The fusing of the first processing result and the second processing result to obtain the second feature information includes:

[0097] Perform global pooling on the first processing result to obtain a global feature vector, and use a fully connected layer and a softmax function to generate a first attention weight corresponding to the first processing result;

[0098] Perform global pooling on the second processing result to obtain a global feature vector, and use a fully connected layer and a softmax function to generate a second attention weight corresponding to the second processing result;

[0099] The second feature information is determined based on the first processing result, the first attention weight, the second processing result and the second attention weight.

[0100] See also Figure 5 , Figure 5 This is a schematic diagram of the detail enhancement module structure in an embodiment of the present application. Figure 5 The detail enhancement module includes: three parallel dilated convolutions 501 with different dilation rates, a connector 502, a first convolution layer, a second convolution layer, a third convolution layer, a first global pooling, a second global pooling, a first fully connected layer and a softmax function, a second fully connected layer and a softmax function, a first weighted module, a second weighted module and a summation module.

[0101] The first fusion result is input into three parallel dilated convolutions with different dilation rates, where the dilation rates of the three parallel dilated convolutions are 1, 3, and 5, respectively. Feature maps with different receptive fields are obtained through the three parallel dilated convolutions with different dilation rates. The three feature maps with different receptive fields are connected through a connector, and then a convolution operation is performed using the first convolutional layer to obtain a first processing result. In a specific implementation, the first convolutional layer can be a 1×1 convolutional layer, which is used to adjust the number of channels so that the number of channels is restored to the number of channels corresponding to the original image.

[0102] At the same time, the first fusion result is input into the second convolution layer, and after processing, it is input into the third convolution layer to obtain the second processing result. The second and third convolution layers here can be 3×3 convolution layers.

[0103] Input the first processing result into the first global pooling for pooling to obtain a global feature vector, and use the first fully connected layer and the softmax function to generate a first attention weight corresponding to the first processing result;

[0104] The second processing result is input into the second global pooling for pooling to obtain a global feature vector, and the second fully connected layer and the softmax function are used to generate a second attention weight corresponding to the second processing result;

[0105] performing weighting using a first weighting module based on the second attention weight and the second processing result; performing weighting using a second weighting module based on the second attention weight and the second processing result;

[0106] The second feature information is obtained by summing the weighted result of the first weighting module and the weighted result of the second weighting module through the summing module.

[0107] The fusion module is used to fuse the first feature information and the second feature information to obtain a second fusion result; the second fusion result is the three-dimensional volume of the semiconductor device to be inspected corresponding to the two DR images with a vertical relationship.

[0108] The preset three-dimensional volume reconstruction model provided in the embodiments of the present application can achieve both large-scale structure restoration and small-scale detail reconstruction.

[0109] The semiconductor device detection process in the embodiment of the present application is described in detail below with reference to the accompanying drawings.

[0110] See also Figure 6 , Figure 6 This is a schematic diagram of the semiconductor device detection process in the embodiment of this application. The specific steps are:

[0111] Step 601 : acquiring two DR images having a vertical relationship with respect to the semiconductor device to be inspected.

[0112] In a specific embodiment of the present application, during the acquisition of two DR images having a vertical relationship, relevant parameters such as tube voltage, tube current, exposure time, and position during detection are ensured to remain unchanged.

[0113] The collected DR images can be processed according to the processing process of DR images during model training, such as size cropping, logarithm taking, normalization, etc., which can better ensure the acquisition of high-precision three-dimensional volumes.

[0114] Step 602 : Obtain the three-dimensional volume of the semiconductor device to be inspected corresponding to two DR images having a perpendicular relationship based on a preset three-dimensional volume reconstruction model.

[0115] Step 603 : performing defect detection on the semiconductor device to be inspected based on the three-dimensional volume.

[0116] In the embodiment of the present application, only by collecting two DR images of the semiconductor device to be inspected with a vertical relationship, the three-dimensional volume of the semiconductor device to be inspected can be obtained using a preset three-dimensional volume reconstruction model, and then defect detection of the semiconductor device can be performed; wherein the preset three-dimensional volume reconstruction model is capable of performing large-scale information recovery and small-scale information reconstruction, and fusing the two results to obtain a high-precision three-dimensional volume. In the embodiment of the present application, only by collecting two DR images of the semiconductor device to be inspected, the three-dimensional volume of the semiconductor device to be inspected can be obtained, so that defect detection of the semiconductor device can be efficiently achieved while saving manpower and material resources; when obtaining the three-dimensional volume of the semiconductor device to be inspected, large-scale information recovery and small-scale information reconstruction are taken into account, so that defect detection of the semiconductor device can be achieved with high precision while saving manpower and material resources.

[0117] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present disclosure, and will not be described in detail here.

[0118] Based on the same inventive concept, the present application also provides a semiconductor device detection device. Figure 7 , Figure 7 This is a schematic diagram of the structure of a semiconductor device detection device in an embodiment of the present application. The device includes:

[0119] An acquisition unit 701 is configured to acquire two DR images having a vertical relationship with respect to the semiconductor device to be inspected;

[0120] An acquisition unit 702 is configured to obtain, based on a preset three-dimensional volume reconstruction model, a three-dimensional volume of the semiconductor device to be inspected corresponding to two DR images having a vertical relationship; wherein the preset three-dimensional volume reconstruction model includes: a channel number enhancement module, a structure recovery module, a detail enhancement module, and a fusion module; the channel number enhancement module is configured to perform channel number enhancement on the two DR images having a vertical relationship while ensuring that the spatial resolution remains unchanged; and to fuse the two DR images after the channel number enhancement to obtain a first fusion result; the structure recovery module is configured to perform large-scale structure recovery on the first fusion result to obtain first feature information; the detail enhancement module is configured to perform small-scale detail reconstruction on the first fusion result to obtain second feature information; and the fusion module is configured to fuse the first feature information with the second feature information to obtain a second fusion result; the second fusion result is the three-dimensional volume of the semiconductor device to be inspected corresponding to the two DR images having a vertical relationship;

[0121] A detection unit 703 is configured to perform defect detection on the semiconductor device to be detected based on the three-dimensional volume;

[0122] In another embodiment, the apparatus further comprises:

[0123] A training unit 704 is configured to generate a 3D volume reconstruction model to be trained; obtain a training data set; wherein the training data set includes multiple data pairs, each data pair including two DR images with a vertical relationship and a 3D volume; the DR image data with a vertical relationship are obtained from a data set consisting of DR images of a reference semiconductor device acquired axially, and the 3D volume is obtained by performing FDK reconstruction based on the data set; and the 3D volume reconstruction model to be trained is trained based on the training data set to obtain a preset 3D volume reconstruction model.

[0124] In another embodiment,

[0125] The training unit 704 is further used to introduce a preset loss function when training the 3D volume reconstruction model to be trained based on the training data set; when the values of N consecutive preset loss functions are all less than the preset loss threshold, the training of the 3D volume reconstruction model to be trained is terminated to obtain the preset 3D volume reconstruction model; wherein N is an integer greater than 1; the preset loss function is a weighted sum of square difference loss and structural index loss; the square difference loss is obtained by calculating the sum of squares of errors between the reference 3D volume and the 3D volume obtained by the 3D volume reconstruction model to be trained; the structural index loss is determined based on the reference 3D volume and the 3D volume obtained by the 3D volume reconstruction model to be trained.

[0126] In another embodiment,

[0127] The structure recovery module uses the encoder to upsample the first fusion result to obtain potential compact representation features; uses the decoder to receive the feature information output by the encoder, reconstructs the potential information through upsampling operations, and obtains the first feature information; and introduces jump connections at different scales in the encoding and decoding process.

[0128] In another embodiment,

[0129] The detail enhancement module obtains feature maps with different receptive fields through three parallel dilated convolutions with different dilation rates on the first fusion result, connects the feature maps with different receptive fields, and uses the convolution layer to restore the number of channels corresponding to the connection result to obtain the first processing result; obtains the second processing result through two convolution layers on the first fusion result; and fuses the first processing result and the second processing result to obtain the second feature information.

[0130] In another embodiment,

[0131] The detail enhancement module fuses the first processing result and the second processing result to obtain the second feature information, including:

[0132] Perform global pooling on the first processing result to obtain a global feature vector, and use a fully connected layer and a softmax function to generate a first attention weight corresponding to the first processing result;

[0133] Perform global pooling on the second processing result to obtain a global feature vector, and use a fully connected layer and a softmax function to generate a second attention weight corresponding to the second processing result;

[0134] The second feature information is determined based on the first processing result, the first attention weight, the second processing result and the second attention weight.

[0135] The units in the above embodiments may be integrated into one body or deployed separately; they may be combined into one unit or further divided into multiple sub-units.

[0136] In another embodiment, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the semiconductor device detection method is implemented when the processor executes the program.

[0137] In another embodiment, a computer-readable storage medium is provided, on which computer instructions are stored. When the instructions are executed by a processor, a semiconductor device detection method is implemented.

[0138] Figure 8 Schematic diagram of the physical structure of the electronic device provided by the embodiment of the present invention. Figure 8 As shown, the electronic device may include: a processor (Processor) 810, a communication interface (Communications Interface) 820, a memory (Memory) 830 and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the following method:

[0139] Acquire two DR images with a vertical relationship for the semiconductor device to be inspected;

[0140] Obtaining the three-dimensional volume of the semiconductor device to be inspected corresponding to the two DR images having a perpendicular relationship based on a preset three-dimensional volume reconstruction model;

[0141] Defect detection of semiconductor devices to be inspected based on three-dimensional volume;

[0142] The preset 3D volume reconstruction model includes: channel number enhancement module, structure recovery module, detail enhancement module and fusion module;

[0143] The channel number enhancement module is used to enhance the channel number of two DR images with a vertical relationship while ensuring that the spatial resolution remains unchanged; and fuse the two DR images after the channel number enhancement to obtain a first fusion result;

[0144] The structure recovery module is used to perform large-scale structure recovery on the first fusion result to obtain first feature information;

[0145] The detail enhancement module is used to reconstruct small-scale details of the first fusion result to obtain second feature information;

[0146] The fusion module is used to fuse the first feature information and the second feature information to obtain a second fusion result; the second fusion result is the three-dimensional volume of the semiconductor device to be inspected corresponding to the two DR images with a vertical relationship.

[0147] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the 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 and includes several instructions for enabling 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 methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0148] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0149] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiment.

[0150] The flowcharts and block diagrams in the accompanying drawings of the present application show the possible implementation architecture, functions and operations of the systems, methods and computer program products according to the various embodiments disclosed in the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in the order of the standards in different figures. For example, the boxes represented by two connections can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of the boxes in the block diagram or flowchart, can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0151] Those skilled in the art will appreciate that the features described in the various embodiments and / or claims disclosed in this application may be combined and / or coupled in various ways, even if such combinations and / or couplings are not explicitly disclosed in this application. In particular, without departing from the spirit and teachings of this application, the features described in the various embodiments and / or claims of this application may be combined and / or coupled in various ways, and all such combinations and / or couplings fall within the scope disclosed in this application.

[0152] The principles and implementation methods of the present invention are described herein using specific embodiments. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas, and is not intended to limit this application. For those skilled in the art, changes can be made in the specific implementation methods and application scope based on the ideas, spirit and principles of the present invention. Any modifications, equivalent replacements, improvements, etc. made therein should be included within the scope of protection of this application.

Claims

1. A semiconductor device detection method, characterized in that: The method comprises: Acquire two DR images with a vertical relationship for the semiconductor device to be inspected; Obtaining a three-dimensional volume of the semiconductor device to be inspected corresponding to the two DR images having a perpendicular relationship based on a preset three-dimensional volume reconstruction model; performing defect detection on the semiconductor device to be inspected based on the three-dimensional volume; The preset 3D volume reconstruction model includes: a channel number enhancement module, a structure recovery module, a detail enhancement module and a fusion module; The channel number increasing module is used to increase the channel number of the two DR images having a vertical relationship while ensuring that the spatial resolution remains unchanged; and to fuse the two DR images after the channel number increase to obtain a first fusion result; The structure recovery module is used to perform large-scale structure recovery on the first fusion result to obtain first feature information; The detail enhancement module is used to reconstruct small-scale details of the first fusion result to obtain second feature information; The fusion module is used to fuse the first feature information and the second feature information to obtain a second fusion result; wherein the second fusion result is a three-dimensional volume of the semiconductor device to be inspected corresponding to the two DR images having a vertical relationship; The structure recovery module uses an encoder to upsample the first fusion result to obtain a potentially compact representation feature; uses a decoder to receive feature information output by the encoder, reconstructs the potential information through an upsampling operation, and obtains the first feature information; and introduces skip connections at different scales during the encoding and decoding process; The detail enhancement module obtains feature maps with different receptive fields through three parallel dilated convolutions with different dilation rates on the first fusion result, connects the feature maps with different receptive fields, and uses a convolution layer to restore the number of channels corresponding to the connection result to obtain a first processing result; inputs the first fusion result into the second convolution layer for processing, and inputs the processed result into the third convolution layer for processing, and obtains the second processing result after processing by the third convolution layer; wherein, the second convolution layer and the third convolution layer are both 3×3 convolution layers; and fuses the first processing result and the second processing result to obtain second feature information.

2. The method according to claim 1, characterized in that The generation of the preset three-dimensional volume reconstruction model includes: Generate a 3D volume reconstruction model to be trained; Acquire a training dataset; wherein the training dataset includes multiple data pairs, each data pair including two DR images with a perpendicular relationship and a reference three-dimensional volume; the DR image data with a perpendicular relationship are acquired from an acquisition dataset consisting of axially acquired DR images of a reference semiconductor device, and the reference three-dimensional volume is obtained by performing FDK reconstruction based on the acquisition dataset; The three-dimensional volume reconstruction model to be trained is trained based on the training data set to obtain a preset three-dimensional volume reconstruction model.

3. The method according to claim 2, characterized in that When training the three-dimensional volume reconstruction model to be trained based on the training data set, introducing a preset loss function; When N consecutive values of the preset loss function are all less than a preset loss threshold, the training of the to-be-trained three-dimensional volume reconstruction model is terminated to obtain a preset three-dimensional volume reconstruction model; Wherein, N is an integer greater than 1; the preset loss function is a weighted sum of a squared error loss function and a structural exponential loss function; the value of the preset loss function is calculated based on the reference three-dimensional volume and the three-dimensional volume obtained by the three-dimensional volume reconstruction model to be trained using the preset loss function.

4. The method according to claim 1, wherein The detail enhancement module fusing the first processing result and the second processing result to obtain second feature information includes: Performing global pooling on the first processing result to obtain a global feature vector, and generating a first attention weight corresponding to the first processing result using a fully connected layer and a softmax function; Performing global pooling on the second processing result to obtain a global feature vector, and generating a second attention weight corresponding to the second processing result using a fully connected layer and a softmax function; Second feature information is determined based on the first processing result, the first attention weight, the second processing result and the second attention weight.

5. A semiconductor device detection device, characterized in that: The device comprises: An acquisition unit, configured to acquire two DR images having a vertical relationship with respect to the semiconductor device to be inspected; An acquisition unit is used to obtain the three-dimensional volume of the semiconductor device to be detected corresponding to the two DR images with a vertical relationship based on a preset three-dimensional volume reconstruction model; wherein the preset three-dimensional volume reconstruction model includes: a channel number improvement module, a structure recovery module, a detail enhancement module and a fusion module; the channel number improvement module is used to respectively improve the channel number of the two DR images with a vertical relationship while ensuring that the spatial resolution remains unchanged; and the two DR images after the channel number improvement are fused to obtain a first fusion result; the structure recovery module is used to perform large-scale structure recovery on the first fusion result to obtain first feature information; the detail enhancement module is used to perform small-scale detail reconstruction on the first fusion result to obtain second feature information; the fusion module is used to fuse the first feature information and the second feature information to obtain a second fusion result; the second fusion result is the semiconductor device to be detected corresponding to the two DR images with a vertical relationship The three-dimensional volume of the device; wherein, the structure recovery module uses the encoder to perform upsampling learning on the first fusion result to obtain potential compact representation features; uses the decoder to receive the feature information output by the encoder, reconstructs the potential information through upsampling operation, and obtains the first feature information; and introduces jump connections at different scales in the encoding and decoding process; the detail enhancement module obtains feature maps with different receptive fields through three parallel dilated convolutions with different dilation rates on the first fusion result, connects the feature maps with different receptive fields, and uses the convolution layer to restore the number of channels corresponding to the connection result to obtain the first processing result; inputs the first fusion result into the second convolution layer for processing, and inputs the processed result into the third convolution layer for processing, and obtains the second processing result after processing by the third convolution layer; wherein, the second convolution layer and the third convolution layer are both 3×3 convolution layers; the first processing result and the second processing result are fused to obtain the second feature information; A detection unit is used to perform defect detection on the semiconductor device to be detected based on the three-dimensional volume.

6. The device according to claim 5, characterized in that The device further comprises: A training unit is configured to generate a three-dimensional volume reconstruction model to be trained; obtain a training data set; wherein the training data set includes multiple data pairs, each data pair including two DR images with a vertical relationship and a three-dimensional volume; the DR image data with a vertical relationship are obtained from a data set consisting of DR images of a reference semiconductor device acquired axially, and the three-dimensional volume is obtained by FDK reconstruction based on the data set; the three-dimensional volume reconstruction model to be trained is trained based on the training data set to obtain a preset three-dimensional volume reconstruction model.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method according to any one of claims 1 to 4 is implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images