Defect data fusion method, device and equipment, storage medium and defect detection method

Abstract

The invention provides a defect data fusion method and device, equipment and a storage medium, and a defect detection method. The method comprises the following steps: obtaining a plurality of target defect data of each target channel in a plurality of target channels and optical characteristics of each target channel; determining at least one fusion channel pair with matched characteristics from the plurality of target channels according to the optical characteristics of each target channel; on the basis that the feature-matched target channels form a fusion channel pair, defect fusion is carried out on multiple pieces of target defect data belonging to different target channels in the fusion channel pair, fusion defect data of the fusion channel pair is obtained, and the fusion defect data is obtained through fusion based on the target defect data of the feature-matched target channels; therefore, accurate defect data can be obtained, unification of defect detection results is achieved, and the requirement for accuracy of defect detection is met.

Description

Technical Field

[0001] The present invention relates to the field of detection, and in particular to a defect data fusion method, device, equipment, storage medium and a defect detection method. Background Art

[0002] In the current field of defect detection, dark-field inspection can be used to detect defects on objects. The principle of dark-field inspection is to illuminate the surface of the object with a beam of light. A detector collects the scattered light from the surface and converts it into an image signal. Because the reflected light from the surface is avoided, only the scattered light at the defect reaches the detector. As a result, the defect appears brighter, while other undefected areas appear darker. The brightness of the image signal can be used to detect the defect location. Dark-field inspection is highly sensitive to scattered light and can be applied to detect tiny surface defects on objects.

[0003] The current dark field detection method usually sets up multiple detectors to form multiple channels for defect detection. However, the defect data obtained by each of the multiple channels is different, resulting in inconsistent defect detection results and failing to meet the accuracy requirements of defect detection. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a defect data fusion method, device, equipment, storage medium and a defect detection method, which can fuse the defect data obtained from each channel detection in multiple channels to obtain fused defect data and achieve the unification of defect detection results.

[0005] To achieve the above objectives, this application has the following technical solutions:

[0006] This application provides a defect data fusion method, the method comprising:

[0007] Acquire a plurality of target defect data of each target channel among a plurality of target channels and an optical characteristic of each target channel;

[0008] determining at least one fusion channel pair with matching characteristics from the plurality of target channels according to the optical characteristics of each target channel;

[0009] Defect fusion is performed on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair.

[0010] Optionally, the optical characteristic comprises a plurality of optical dimensions, the plurality of optical dimensions comprising a first optical dimension and a second optical dimension, and defect variability in the first optical dimension is smaller than defect variability in the second optical dimension;

[0011] Determining at least one fusion channel pair with matching characteristics from the multiple target channels according to the optical characteristics of each target channel includes:

[0012] determining, based on the first optical dimension, a first fused channel pair from the plurality of target channels that matches at least one characteristic of the first optical dimension;

[0013] fusing the first fusion channel pair that performs characteristic matching on the first optical dimension to obtain a first-order fusion channel;

[0014] determining, based on the second optical dimension, from the at least one first-order fused channel, a second fused channel pair having at least one matching characteristic in the second optical dimension;

[0015] The second fusion channel pair that performs characteristic matching on the second optical dimension is fused to obtain a second-order fusion channel.

[0016] Optionally, the multiple optical dimensions further include a third optical dimension; and the method further includes:

[0017] Based on the third optical dimension, determining at least one third fused channel pair with characteristic matching in the third optical dimension from at least one second-order fused channel in the second optical dimension; fusing the third fused channel pair with characteristic matching in the third optical dimension to obtain a third-order fused channel;

[0018] or,

[0019] Based on the third optical dimension, at least one third fusion channel pair with characteristic matching in the third optical dimension is determined from the at least one first-order fusion channel; and the third fusion channel pair with characteristic matching in the third optical dimension is fused to obtain a third-order fusion channel.

[0020] Optionally, performing defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair includes:

[0021] Performing defect fusion on a plurality of target defect data belonging to different target channels in the first fusion channel pair to obtain first fused defect data of the first-order fusion channel;

[0022] performing defect fusion on first fused defect data belonging to different first-order fusion channels in the second fusion channel pair to obtain second fused defect data of the second-order fusion channel;

[0023] Defect fusion is performed on the second fused defect data belonging to different second-order fusion channels in the third fusion channel pair to obtain third fused defect data of the third-order fusion channel.

[0024] Optionally, performing defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair includes:

[0025] Performing defect fusion on a plurality of target defect data belonging to different target channels in the first fusion channel pair to obtain first fused defect data of the first-order fusion channel;

[0026] Defect fusion is performed on the first fused defect data belonging to different first-order fusion channels in the third fusion channel pair to obtain third fused defect data of the third-order fusion channel.

[0027] Optionally, the optical characteristics include channel sensitivity, channel width and light source type.

[0028] Optionally, the plurality of target defect data include defect type data, defect attribute data and defect location data; the fused defect data include type fusion data, attribute fusion data and location fusion data;

[0029] The step of performing defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair includes:

[0030] Performing type fusion on a plurality of defect type data belonging to different target channels in the fusion channel pair to obtain type fusion data of the fusion channel pair;

[0031] Performing attribute fusion on a plurality of defect attribute data belonging to different target channels in the fusion channel pair to obtain attribute fusion data of the fusion channel pair;

[0032] Position fusion is performed on a plurality of defect position data belonging to different target channels in the fused channel pair to obtain position fusion data of the fused channel pair.

[0033] Optionally, performing type fusion on a plurality of defect type data belonging to different target channels in the fused channel pair to obtain type fusion data of the fused channel pair includes:

[0034] If the multiple defect type data belonging to different target channels in the fused channel pair are of the same defect type, the defect type of the type fused data of the fused channel pair and the multiple defect type data of the target channel are the same;

[0035] If the multiple defect type data belonging to different target channels in the fusion channel pair are different defect types, type fusion is performed on the multiple defect type data of different defect types according to a preset type fusion condition to obtain type fusion data of the fusion channel pair.

[0036] Optionally, performing position fusion on a plurality of defect position data respectively belonging to different target channels in the fused channel pair to obtain position fusion data of the fused channel pair includes:

[0037] Obtaining coordinate weight values ​​of different target channels in the fused channel pair;

[0038] Position fusion data of the plurality of defect positions is performed according to a preset coordinate fusion condition and the coordinate weight value to obtain position fusion data of the fusion channel pair.

[0039] Optionally, the preset coordinate fusion condition is to determine the defect position data with the largest coordinate weight value;

[0040] The performing position fusion on the plurality of defect position data according to the preset coordinate fusion condition and the coordinate weight value to obtain the position fusion data of the fusion channel pair includes:

[0041] Determine the defect position data with the largest coordinate weight value from the coordinate weight values ​​of the different target channels as the position fusion data of the fusion channel pair.

[0042] Optionally, performing attribute fusion on a plurality of defect attribute data belonging to different target channels in the fusion channel pair to obtain attribute fusion data of the fusion channel pair includes:

[0043] Determining a fusion classification algorithm for the fusion channel pair according to optical characteristics of a target channel included in the fusion channel pair;

[0044] Attribute fusion is performed on a plurality of attribute defect data belonging to different target channels in the fusion channel pair according to the type defect data and the fusion classification algorithm of the fusion channel pair to obtain attribute fusion data of the fusion channel pair.

[0045] Optionally, the plurality of target defect data include defect location data;

[0046] The step of performing defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair includes:

[0047] Clustering the defect position data belonging to different target channels in the fused channel pair to obtain a plurality of clusters belonging to different target channels;

[0048] Matching the defect location data included in the plurality of clusters respectively belonging to different target channels to obtain matched defect data respectively belonging to the different target channels;

[0049] Defect fusion is performed on the matching defect data belonging to different target channels to obtain fused defect data of the fused channel pair.

[0050] The present application provides a defect detection method, according to any one of the above-mentioned defect data fusion methods, the defect detection method comprising:

[0051] Acquiring fusion defect data of the fusion channel pair;

[0052] The defect type, defect attribute, and defect location of a target defect corresponding to the fused defect data are determined according to the fused defect data.

[0053] The present application provides a defect data fusion device, comprising:

[0054] an acquiring unit, configured to acquire a plurality of target defect data of each target channel among a plurality of target channels and an optical characteristic of each target channel;

[0055] a determining unit, configured to determine at least one fusion channel pair with matching characteristics from the multiple target channels according to the optical characteristics of each target channel;

[0056] The fusion unit is configured to perform defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair.

[0057] The present application provides a defect data fusion device, the device comprising: a processor and a memory;

[0058] The memory is used to store instructions;

[0059] The processor is configured to execute the instructions in the memory and perform any of the methods described above.

[0060] The present application provides a computer storage medium, characterized in that the computer storage medium is used to store a computer program, and when the computer program is run on a computer device, the computer device executes any one of the above methods.

[0061] The present application provides a defect data fusion method, which includes: obtaining multiple target defect data of each target channel in multiple target channels and the optical characteristics of each target channel, wherein the optical characteristics of each target channel reflect the characteristics of the target defect data collected by the target channel, so as to realize subsequent defect fusion of the target defect data of different target channels based on the optical characteristics of the target channel; determining at least one characteristic-matched fusion channel pair from the multiple target channels according to the optical characteristics of each target channel, that is, characteristic matching is performed on the multiple target channels based on the optical characteristics of different target channels, and the characteristics of the target defect data of the characteristic-matched target channels are also matched, which is conducive to defect fusion; on the basis of forming a fusion channel pair based on the characteristic-matched target channels, defect fusion is performed on multiple target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair, wherein the fused defect data is obtained by fusion of the target defect data of the characteristic-matched target channels, so that accurate defect data can be obtained, the unification of defect detection results is achieved, and the accuracy requirements of defect detection are met. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0063] Figure 1 A schematic diagram of a defect data fusion method provided in an embodiment of the present application is shown;

[0064] Figure 2 A schematic diagram of the structure of multiple target channels provided in an embodiment of the present application is shown;

[0065] Figure 3 A schematic diagram of channel fusion provided by an embodiment of the present application is shown;

[0066] Figure 4 A schematic diagram of a defect detection method according to an embodiment of the present invention is shown;

[0067] Figure 5 A schematic structural diagram of a defect data fusion device provided in an embodiment of the present application is shown;

[0068] Figure 6 A structural schematic diagram of a defect detection device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0069] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below with reference to the accompanying drawings.

[0070] In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0071] This application is described in detail with reference to schematic diagrams. When describing the embodiments of this application, for ease of explanation, cross-sectional views of device structures may be partially enlarged and not to scale. Furthermore, these schematic diagrams are merely illustrative and should not limit the scope of protection of this application. Furthermore, in actual production, three-dimensional dimensions, including length, width, and depth, should be included.

[0072] Wafer surface defect detection is a crucial part of the semiconductor manufacturing process. Its main purpose is to detect defects on the wafer surface to ensure the quality of chip manufacturing and the performance of the final product.

[0073] Darkfield inspection is a method used to detect surface defects on wafers. The principle of darkfield inspection is to illuminate the wafer surface with a beam of light, collect the scattered light from the wafer surface using a detector, and convert this scattered light into an image signal. Because the reflected light from the wafer surface is avoided, only the scattered light at the defect reaches the detector. As a result, the defect appears brighter, while other defect-free areas appear darker. The defect location can be detected by the brightness and darkness of the image signal. Darkfield inspection is highly sensitive to scattered light and can be applied to detect tiny surface defects on wafers.

[0074] The current dark field detection method usually sets up multiple detectors to form multiple channels for defect detection. However, the defect data obtained by each channel in the multiple channels is different. For example, the defect position data obtained by each channel may be biased, resulting in defect duplication during defect detection. That is, different channels detect the same defect as different defects, and the defect attribute data and defect type data obtained by different channels for the same defect may also be different, which ultimately leads to inconsistent defect detection results and cannot meet the accuracy requirements of defect detection.

[0075] Based on this, the present application provides a defect data fusion method, which includes: obtaining multiple target defect data of each target channel in multiple target channels and the optical characteristics of each target channel, the optical characteristics of each target channel reflecting the characteristics of the target defect data collected by the target channel, so as to realize subsequent defect fusion of the target defect data of different target channels based on the optical characteristics of the target channel; determining at least one characteristic-matched fusion channel pair from multiple target channels according to the optical characteristics of each target channel, that is, characteristic matching is performed on multiple target channels based on the optical characteristics of different target channels, and the characteristics of the target defect data of the characteristic-matched target channels are also matched, which is conducive to defect fusion; on the basis of forming a fusion channel pair based on the characteristic-matched target channels, defect fusion is performed on multiple target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fused channel pair, and the fused defect data is obtained by fusion of the target defect data of the characteristic-matched target channels, so that accurate defect data can be obtained, the unification of defect detection results is achieved, and the accuracy requirements of defect detection are met.

[0076] In order to better understand the technical solutions and technical effects of the present application, specific embodiments will be described in detail below with reference to the accompanying drawings.

[0077] refer to Figure 1 FIG. 1 is a flow chart of a defect data fusion method provided in an embodiment of the present application, the method comprising the following steps:

[0078] S101 , acquiring a plurality of target defect data of each target channel among a plurality of target channels and an optical characteristic of each target channel.

[0079] In the embodiment of the present application, the channel includes an incident channel and a collection channel, and the target channel is a collection channel. The incident channel is used for incident light, and the collection channel is used to collect scattered light reflected by the target defect and form target defect data, that is, each target channel collects target defect data of the target defect, so multiple target defect data of each target channel in multiple target channels can be obtained, thereby providing a data detection basis for the target defect. Specifically, the incident channel includes a normal incident channel (Normal) and an oblique incident channel (Oblique), refer to Figure 2 shown.

[0080] Due to the different optical characteristics of the incident channels and the different physical properties of the target channels, the optical characteristics of the target channels are different. The characteristics of the target defect data collected by target channels with different optical characteristics are also different. The optical characteristics of each target channel reflect the characteristics of the target defect data collected by the target channel, enabling subsequent defect fusion of target defect data from different target channels based on the optical characteristics of the target channels.

[0081] S102: Determine at least one fusion channel pair with matching characteristics from multiple target channels according to the optical characteristics of each target channel.

[0082] In an embodiment of the present application, after obtaining the optical characteristics of each target channel, since the target defect data of the same target defect collected from different target channels will be subsequently fused, at least one characteristic-matched fusion channel pair can be determined from multiple target channels based on the optical characteristics of each target channel, that is, characteristic matching is performed on multiple target channels based on the optical characteristics of different target channels, and the characteristics of the target defect data of the characteristic-matched target channels are also matched, which is conducive to defect fusion.

[0083] Specifically, a feature-matched fusion channel pair includes two feature-matched target channels, i.e., two feature-matched target channels are determined from multiple target channels. Since there are multiple target channels, at least one set of two feature-matched target channels can be determined from the multiple target channels, i.e., at least one feature-matched fusion channel pair can be determined from the multiple target channels.

[0084] S103 , performing defect fusion on a plurality of target defect data belonging to different target channels in the fused channel pair to obtain fused defect data of the fused channel pair.

[0085] In an embodiment of the present application, on the basis of forming a fusion channel pair based on the target channel of feature matching, defect fusion is performed on multiple target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair. The fused defect data is obtained by fusing the target defect data of the same target defect based on the target channel of feature matching. Therefore, accurate defect data can be obtained, the unification of defect detection results can be achieved, and the accuracy requirements of defect detection can be met.

[0086] In an embodiment of the present application, the optical characteristics include channel sensitivity, channel width and light source type. Channel sensitivity includes high sensitivity and low sensitivity. Channel width includes wide channel and narrow channel. Channel width is the collection angle of scattered light. The collection angle of wide channel is larger, and the collection angle of narrow channel is smaller. Light source type includes oblique incident light source and vertical incident light source. Based on different optical characteristics, target channels can be divided into multiple types, as shown in Table 1, which shows the channel names and channel optical characteristics of 8 target channels based on different optical characteristics.

[0087] Table 1

[0088]

[0089]

[0090] In an embodiment of the present application, the optical characteristics include multiple optical dimensions. Therefore, when determining a fused channel pair from multiple target channels based on characteristic matching of the optical characteristics of each target channel, the fused channel pair is determined based on characteristic matching in different optical dimensions, thereby realizing determination of the fused channel pair based on characteristic matching using optical dimensions as the basis.

[0091] Specifically, the multiple optical dimensions include a first optical dimension and a second optical dimension. The difference between the two optical dimensions lies in the defect variability. The defect variability of the first optical dimension is smaller than the defect variability of the second optical dimension. That is, when the target channels of the two optical dimensions collect defect data for the same target defect, the defect variability between the target defect data corresponding to the first optical dimension is smaller than the defect variability between the target defect data corresponding to the second optical dimension.

[0092] As an example, the first optical dimension is channel sensitivity, and the second optical dimension is channel width, that is, the defect difference between different channel sensitivities is smaller than the defect difference between different channel widths, and the defect difference between target defect data corresponding to different channel sensitivities is smaller than the defect difference between target defect data corresponding to different channel widths.

[0093] Considering that optical characteristics include multiple optical dimensions, channel fusion can be performed on the target channels based on the optical dimensions, and then data fusion can be performed on the target defect data collected by the target channels. Channel fusion can be performed on the target channels in the optical dimensions with smaller defect differences first, and then the channel fusion of the target channels in the optical dimensions with larger defect differences can be continued. That is, data fusion of the target defect data corresponding to the optical dimensions with smaller defect differences is performed first, and then the data fusion of the target defect data corresponding to the optical dimensions with larger defect differences is continued, which helps to improve the accuracy of the target defect data during data fusion.

[0094] Specifically, when channel fusion is performed on target channels based on optical dimensions, a first fusion channel pair that matches at least one characteristic of the first optical dimension is determined from multiple target channels based on the first optical dimension; the first fusion channel pair that matches the characteristics of the first optical dimension is fused to obtain a first-order fusion channel; a second fusion channel pair that matches at least one characteristic of the second optical dimension is determined from at least one first-order fusion channel based on the second optical dimension; and the second fusion channel pair that matches the characteristics of the second optical dimension is fused to obtain a second-order fusion channel. In other words, first, characteristic matching and channel fusion of the target channels are performed based on the first optical dimension to obtain a first-order fusion channel, and then characteristic matching and channel fusion of the first-order fusion channels are performed based on the second optical dimension to obtain a second-order fusion channel, i.e., channel fusion is performed at different stages based on different optical dimensions, thereby gradually fusion of multiple target channels, which in turn helps to gradually fuse multiple target defect data, ultimately obtaining accurate defect data and achieving unified defect detection results.

[0095] As an example, the first optical dimension is channel sensitivity, and the second optical dimension is channel width. Based on the channel sensitivity, first fused channel pairs with high and low channel sensitivities are determined from multiple target channels; the first fused channel pairs with high and low channel sensitivities are fused to obtain a first-order fused channel; based on the channel width, second fused channel pairs with wide and narrow channel widths are determined from at least one first-order fused channel; the second fused channel pairs with wide and narrow channel widths are fused to obtain a second-order fused channel.

[0096] In an embodiment of the present application, the multiple optical dimensions include not only the first optical dimension and the second optical dimension, but also the third optical dimension. When channel fusion is performed on the target channel based on the optical dimension, the first optical dimension, the second optical dimension and the third optical dimension can be used for three-way channel fusion, or the first optical dimension and the third optical dimension can be used for two-way channel fusion, thereby meeting different channel fusion and data fusion scenarios in actual situations.

[0097] As a possible implementation method, based on the third optical dimension, at least one third fusion channel pair with matching characteristics in the third optical dimension is determined from at least one second-order fusion channel in the second optical dimension, and the third fusion channel pair with matching characteristics in the third optical dimension is fused to obtain a third-order fusion channel. That is, after performing secondary channel fusion using the first and second optical dimensions, the third optical dimension is continued to be used for the third channel fusion.

[0098] As an example, the first optical dimension is channel sensitivity, the second optical dimension is channel width, and the third optical dimension is light source type. Figure 3As shown in FIG, multiple target channels are DWLO, DWHO, DNLO, DNHO, DWLN, DWHN, DNLN, and DNHN according to the three optical dimensions. Based on the channel sensitivity, the first fusion channel pairs with high and low channel sensitivity are determined to be DWLO and DWHO, DNLO and DNHO, DWLN and DWHN, and DNLN and DNHN, respectively. DWLO and DWHO, DNLO and DNHO, DWLN and DWHN, and DNLN and DNHN are fused respectively to obtain first-order fusion channels DCWWO, DCNNO, DCWWN, and DCNNN, as shown in Table 2, which shows the channel names and channel optical characteristics of the fused channels. Based on channel width, the second fusion channel pairs with wide and narrow channel widths are determined from DCWWO, DCNNO, DCWWN, and DCNNN, respectively. DCWWO and DCNNO, DCWWN and DCNNN are fused separately to obtain second-order fusion channels DCO and DCN. Data fusion of the target defect data corresponding to DCO and DCN yields accurate defect data. Furthermore, based on light source type, the third fusion channel pair with oblique and vertical incident light sources is determined to be DCO and DCN. DCO and DCN are fused to obtain a third-order fusion channel, DC.

[0099] Table 2

[0100] index Channel Name Channel optical characteristics 1 DCWWO Oblique incidence wide channel high and low sensitivity fusion channel 2 DCNNO Oblique incidence narrow channel high and low sensitivity fusion channel 3 DCWWN Normal incidence wide channel high and low sensitivity fusion channel 4 DCNNN Normal incidence narrow channel high and low sensitivity fusion channel 5 DCO Oblique-incidence fusion channel 6 DCN Normal incidence fusion channel 7 DW Normal oblique incidence wide channel fusion channel 8 DN Normal oblique incidence narrow channel fusion channel 9 DC Normal oblique incidence fusion channel

[0101] As another possible implementation, based on the third optical dimension, at least one third fused channel pair with matching characteristics in the third optical dimension is determined from at least one first-order fused channel; the third fused channel pair with matching characteristics in the third optical dimension is fused to obtain a third-order fused channel, that is, after performing channel fusion once using the first optical dimension, a second channel fusion is performed using the third optical dimension.

[0102] As an example, the first optical dimension is channel sensitivity and the third optical dimension is light source type. Figure 3As shown in FIG, multiple target channels are DWLO, DWHO, DNLO, DNHO, DWLN, DWHN, DNLN, and DNHN according to the three optical dimensions. Based on the channel sensitivity, the first fusion channel pairs with high and low channel sensitivity are determined to be DWLO and DWHO, DNLO and DNHO, DWLN and DWHN, and DNLN and DNHN, respectively. DWLO and DWHO, DNLO and DNHO, DWLN and DWHN, and DNLN and DNHN are fused respectively to obtain first-order fusion channels DCWWO, DCNNO, DCWWN, and DCNNN, as shown in Table 2, which shows the channel names and channel optical characteristics of the fused channels. Based on the light source type, the third fusion channel pairs with oblique incidence and vertical incidence are determined from DCWWO, DCNNO, DCWWN and DCNNN, respectively. DCWWO and DCWWN, DCNNO and DCNNN are fused respectively to obtain the third-order fusion channels DW and DN.

[0103] In an embodiment of the present application, when channel fusion is performed using different optical dimensions, data fusion is also performed simultaneously, thereby achieving the unification of defect data of different target channels.

[0104] As a possible implementation, during the secondary channel fusion process using the first and second optical dimensions, the target defect data is also fused twice. Multiple target defect data belonging to different target channels in the first fusion channel pair are fused to obtain first fused defect data for the first-order fusion channel. First fused defect data belonging to different first-order fusion channels in the second fusion channel pair are fused to obtain second fused defect data for the second-order fusion channel. In other words, the target defect data is gradually fused in two stages to obtain accurate defect data.

[0105] As another possible implementation method, in the process of performing three-way channel fusion using the first optical dimension, the second optical dimension, and the third optical dimension, the target defect data is also fused three times. Defect fusion is performed on multiple target defect data belonging to different target channels in the first fusion channel pair to obtain the first fused defect data of the first-order fusion channel. Defect fusion is performed on the first fused defect data belonging to different first-order fusion channels in the second fusion channel pair to obtain the second fused defect data of the second-order fusion channel. Defect fusion is performed on the second fused defect data belonging to different second-order fusion channels in the third fusion channel pair to obtain the third fused defect data of the third-order fusion channel. In other words, in the process of performing three-time fusion on the defect data, the data fusion of this stage is continued based on the fusion data after the data fusion of the previous stage, thereby realizing the three-stage gradual fusion of the target defect data and finally obtaining accurate defect data.

[0106] As another possible implementation, during the secondary channel fusion process using the first and third optical dimensions, the target defect data is also fused twice. Multiple target defect data belonging to different target channels in the first fusion channel pair are fused to obtain first fused defect data for the first-order fusion channel. First fused defect data belonging to different first-order fusion channels in the third fusion channel pair are fused to obtain third fused defect data for the third-order fusion channel. In other words, the target defect data is gradually fused in two stages to obtain accurate defect data.

[0107] In an embodiment of the present application, multiple target defect data include defect type data, defect attribute data, and defect location data, that is, each target channel collects the defect type data, defect attribute data, and defect location data of the target defect when collecting the target defect data of the target defect. The fused defect data includes type fusion data, attribute fusion data, and location fusion data. When performing defect fusion on multiple target defect data belonging to different target channels in a fusion channel pair, type fusion, attribute fusion, and location fusion are performed to obtain type fusion data, attribute fusion data, and location fusion data, respectively. That is, type fusion is performed on multiple defect type data belonging to different target channels in a fusion channel pair to obtain type fusion data of the fusion channel pair, attribute fusion is performed on multiple defect attribute data belonging to different target channels in a fusion channel pair to obtain attribute fusion data of the fusion channel pair, and location fusion is performed on multiple defect location data belonging to different target channels in a fusion channel pair to obtain location fusion data of the fusion channel pair.

[0108] The following is a detailed introduction to the type fusion of multiple defect type data belonging to different target channels in the fusion channel pair.

[0109] When performing type fusion on multiple defect type data belonging to different target channels in the fused channel pair, it can be determined first whether the defect types of the multiple defect type data belonging to different target channels in the fused channel pair are the same; if the multiple defect type data belonging to different target channels in the fused channel pair are the same defect type, the type fusion data of the fused channel pair is obtained to have the same defect type as the multiple defect type data of the target channel, that is, if the defect types of the multiple defect type data belonging to different target channels in the fused channel pair are the same, the defect type of the type fusion data after type fusion still remains the defect type of the defect type data of the target channel; if the multiple defect type data belonging to different target channels in the fused channel pair are different defect types, the multiple defect type data of different defect types are type fused according to preset type fusion conditions to obtain the type fusion data of the fused channel pair.

[0110] The preset type fusion conditions are the conditions for performing type fusion on defect type data of different defect types. As an example, the preset type fusion conditions are shown in Table 3, which shows the defect types of type fusion data obtained by type fusion when the defect type of the target defect data is slip line (SLIP), large area defect (AREA), or particle defect (LPD / LPDN). The left and top columns of Table 3 show the defect types of multiple defect type data belonging to different target channels in the fused channel pair. The remaining columns in Table 3 show the defect types of the type fusion data.

[0111] Table 3

[0112] Slip Line Large area defects Particle defects Slip Line Slip Line Slip Line Slip Line Large area defects Slip Line Large area defects Large area defects Particle defects Slip Line Large area defects Particle defects

[0113] The following is a detailed introduction to the position fusion of multiple defect position data belonging to different target channels in the fusion channel pair.

[0114] When performing position fusion on multiple defect location data belonging to different target channels in a fused channel pair, the coordinate weight values ​​of the different target channels in the fused channel pair can be obtained. The multiple defect location data are then positionally fused according to the preset coordinate fusion conditions and the coordinate weight values ​​to obtain position fusion data for the fused channel pair. The preset coordinate fusion conditions are the conditions for position fusion of multiple defect location data. In other words, each target channel collects defect location data for the target defect, and the importance of the defect location data of the target channel is measured based on the coordinate weight value of the target channel. The position fusion data is further determined in combination with the preset coordinate fusion conditions.

[0115] As a possible implementation method, the preset coordinate fusion condition is to determine the defect location data with the largest coordinate weight value. When multiple defect location data are positionally fused according to the preset coordinate fusion condition and the coordinate weight value, the defect location data with the largest coordinate weight value is determined from the coordinate weight values ​​of different target channels as the location fusion data of the fusion channel pair. The largest coordinate weight value reflects that the defect location data collected by the corresponding target channel is more important. Determining the defect location data with the largest coordinate weight value as the location fusion data of the fusion channel pair can improve the accuracy of the location fusion data after position fusion.

[0116] The following is a detailed introduction to the attribute fusion of multiple defect attribute data belonging to different target channels in the fusion channel pair.

[0117] When performing attribute fusion on multiple defect attribute data belonging to different target channels in a fused channel pair, the fusion classification algorithm of the fused channel pair can be determined based on the optical characteristics of the target channels included in the fused channel pair, and attribute fusion is performed on multiple attribute defect data belonging to different target channels in the fused channel pair based on the type defect data and the fusion classification algorithm of the fused channel pair. That is, attribute fusion is performed on the attribute defect data by combining the type defect data and the fusion classification algorithm, so as to obtain accurate attribute fusion data of the fused channel pair.

[0118] When determining a matching fusion channel pair based on the optical characteristics of the target channel, a fusion classification algorithm for the fusion channel pair is also determined based on the optical characteristics of the target channel included in the fusion channel pair. Since optical characteristics can include three optical dimensions: channel sensitivity, channel width, and light source type, three fusion classification algorithms are defined when determining fusion channel pairs at different stages based on these three optical dimensions: a high-low sensitivity fusion algorithm, a wide-narrow channel fusion algorithm, and a normal-oblique-incidence fusion algorithm.

[0119] As an example, based on channel sensitivity, the first fusion channel pairs with high and low channel sensitivity are determined to be DWLO and DWHO, DNLO and DNHO, DWLN and DWHN, DNLN and DNHN, respectively, and the fusion classification algorithm of the first fusion channel pair is the high-low sensitivity fusion algorithm; based on the channel width, the second fusion channel pairs are DCWWO and DCNNO, DCWWN and DCNNN, respectively, and the fusion classification algorithm of the second fusion channel pair is the wide and narrow channel fusion algorithm; based on the light source type, the third fusion channel pairs are DCO and DCN, DCWWO and DCWWN, DCNNO and DCNNN, and the fusion classification algorithm of the third fusion channel pair is the normal oblique incident light fusion algorithm.

[0120] When using different fusion classification algorithms to perform attribute fusion on attribute defect data, the defect type of the target defect corresponding to the attribute defect data will also affect the attribute fusion. That is, for the high and low sensitivity fusion algorithm, the wide and narrow channel fusion algorithm, and the normal oblique incident light fusion algorithm, the attribute defect data can be combined with the type defect data of the target defect corresponding to the attribute defect data to perform attribute fusion on the attribute defect data.

[0121] As an example, referring to Table 4, Table 4 shows that attribute fusion data for the fused channel pair can be obtained by performing attribute fusion on multiple attribute defect data belonging to different target channels in the fused channel pair based on the type defect data and the high and low sensitivity fusion algorithm of the fused channel pair. Table 4 shows the defect types of the target defects corresponding to the type defect data, such as slip line (SLIP), large area defect (AREA), unsaturated particle defect, or saturated particle defect. The left and top columns of Table 4 show the defect types of the target defects corresponding to the multiple type defect data belonging to different target channels in the fused channel pair, and the rest of Table 4 shows the attribute fusion data.

[0122] Table 4

[0123]

[0124] As another example, referring to Table 5, Table 5 shows that attribute fusion data for the fused channel pair can be obtained by performing attribute fusion on multiple attribute defect data belonging to different target channels in the fused channel pair based on the type defect data and the wide-narrow channel fusion algorithm of the fused channel pair. Table 5 shows the defect type of the target defect corresponding to the type defect data, which is a slip line defect (SLIP), a large area defect (AREA), an unsaturated particle defect, or a saturated particle defect. The left and top columns of Table 5 show the defect types of the target defects corresponding to the multiple type defect data belonging to different target channels in the fused channel pair, and the rest of Table 5 shows the attribute fusion data.

[0125] Table 5

[0126]

[0127]

[0128] As another example, referring to Table 6, Table 6 shows that attribute fusion data for the fused channel pair can be obtained by performing attribute fusion on multiple attribute defect data belonging to different target channels in the fused channel pair based on the type defect data and the normal oblique incident light fusion algorithm of the fused channel pair. Table 6 shows the defect types of the target defects corresponding to the type defect data, such as slip line (SLIP), large area defect (AREA), unsaturated particle defect, or saturated particle defect. The left and top columns of Table 6 show the defect types of the target defects corresponding to the multiple type defect data belonging to different target channels in the fused channel pair, and the rest of Table 6 shows the attribute fusion data.

[0129] Table 6

[0130]

[0131] In an embodiment of the present application, when defect fusion is performed on multiple target defect data belonging to different target channels in a fusion channel pair, it is necessary to determine whether the multiple target defect data belonging to different target channels in the fusion channel pair correspond to the same target defect, and then defect fusion is performed on the target defect data of the target defect. Therefore, after determining the fusion channel pair with characteristic matching, the defect position data in the multiple target defect data is obtained, the defect position data belonging to different target channels in the fusion channel pair are clustered to obtain multiple cluster clusters belonging to different target channels, the defect position data included in the multiple cluster clusters belonging to different target channels are matched to obtain matching defect data belonging to different target channels, and the matching defect data belonging to different target channels correspond to the same target defect. Subsequently, defect fusion is performed on the matching defect data belonging to different target channels, that is, the matching defect data belonging to different target channels belonging to the same target defect are defect fused, and finally the fused defect data of the fusion channel pair is obtained.

[0132] As an example, the defect center coordinates are calculated for the defect location data belonging to different target channels in the fused channel pair. Based on the defect center coordinates, a preset clustering radius is used to cluster the defect location data belonging to different target channels in the fused channel pair. Cluster clusters are generated according to the distance from the defect center coordinates. The defect location data included in multiple clusters belonging to different target channels are matched using a two-way matching algorithm. The defect location data belonging to different target channels that are successfully matched are determined as matched defect data, and the defect location data that are unsuccessfully matched are determined as independent defect data. Defect fusion is performed on the matched defect data, and the independent defect data is directly output.

[0133] It can be seen that the defect data fusion method provided by the present application obtains multiple target defect data of each target channel in multiple target channels and the optical characteristics of each target channel. The optical characteristics of each target channel reflect the characteristics of the target defect data collected by the target channel, so as to realize the subsequent defect fusion of the target defect data of different target channels based on the optical characteristics of the target channel; according to the optical characteristics of each target channel, at least one characteristic-matched fusion channel pair is determined from the multiple target channels, that is, the multiple target channels are characteristically matched based on the optical characteristics of different target channels, and the characteristics of the target defect data of the characteristic-matched target channels are also matched, which is conducive to defect fusion; on the basis of forming a fusion channel pair based on the characteristic-matched target channels, defect fusion is performed on the multiple target defect data belonging to different target channels in the fusion channel pair to obtain the fused defect data of the fused channel pair. The fused defect data is obtained by fusion of the target defect data of the characteristic-matched target channels, so that accurate defect data can be obtained, the unification of defect detection results can be achieved, and the accuracy requirements of defect detection can be met.

[0134] Based on the defect data fusion method provided in the above embodiment, the present application embodiment also provides a defect detection method, referring to Figure 4 FIG. 1 is a flow chart of a defect detection method provided in an embodiment of the present application. The defect detection method provided in an embodiment of the present application includes the following steps:

[0135] S201, obtaining fusion defect data of a fusion channel pair.

[0136] In an embodiment of the present application, after channel fusion is performed on multiple target channels and defect fusion of target defect data is performed based on the above embodiment, fused defect data of the fused channel pair is obtained to serve as a data basis for target defect detection.

[0137] S202 , determining the defect type, defect attribute, and defect location of a target defect corresponding to the fused defect data according to the fused defect data.

[0138] In an embodiment of the present application, after obtaining the fused defect data, the fused defect data can accurately reflect the situation of the target defect. Therefore, the defect type, defect attributes and defect location of the target defect corresponding to the fused defect data can be determined based on the fused defect data, thereby achieving accurate detection of the target defect.

[0139] Based on the defect data fusion method provided in the above embodiment, the present application embodiment also provides a defect data fusion device, referring to Figure 5 FIG. 5 is a schematic diagram of the structure of a defect data fusion device provided in an embodiment of the present application. The defect data fusion device 500 provided in an embodiment of the present application includes:

[0140] an acquisition unit 510, configured to acquire a plurality of target defect data of each target channel among a plurality of target channels and an optical characteristic of each target channel;

[0141] a determining unit 520, configured to determine at least one fusion channel pair having matching characteristics from a plurality of target channels according to the optical characteristics of each target channel;

[0142] The fusion unit 530 is configured to perform defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair.

[0143] As a possible implementation, the optical characteristic includes multiple optical dimensions, the multiple optical dimensions include a first optical dimension and a second optical dimension, and defect variability in the first optical dimension is smaller than defect variability in the second optical dimension;

[0144] The determining unit 520 is configured to:

[0145] Based on the first optical dimension, determining a first fused channel pair from a plurality of target channels that matches at least one characteristic of the first optical dimension;

[0146] fusing the first fusion channel pair that performs characteristic matching on the first optical dimension to obtain a first-order fusion channel;

[0147] determining, based on the second optical dimension, from the at least one first-order fused channel, a second fused channel pair having at least one characteristic matching the second optical dimension;

[0148] The second fusion channel pair that performs characteristic matching on the second optical dimension is fused to obtain a second-order fusion channel.

[0149] As a possible implementation, the multiple optical dimensions further include a third optical dimension; the device further includes a third-order fusion unit; the third-order fusion unit is configured to:

[0150] Based on the third optical dimension, determining at least one third fused channel pair with matching characteristics in the third optical dimension from at least one second-order fused channel in the second optical dimension; fusing the third fused channel pairs with matching characteristics in the third optical dimension to obtain a third-order fused channel;

[0151] or,

[0152] Based on the third optical dimension, at least one third fusion channel pair with characteristic matching in the third optical dimension is determined from at least one first-order fusion channel; the third fusion channel pair with characteristic matching in the third optical dimension is fused to obtain a third-order fusion channel.

[0153] As a possible implementation, the fusion unit 530 is configured to:

[0154] Perform defect fusion on a plurality of target defect data belonging to different target channels in the first fusion channel pair to obtain first fused defect data of a first-order fusion channel;

[0155] Perform defect fusion on the first fused defect data belonging to different first-order fusion channels in the second fusion channel pair to obtain second fused defect data of the second-order fusion channel;

[0156] Defect fusion is performed on the second fused defect data belonging to different second-order fusion channels in the third fusion channel pair to obtain third fused defect data of the third-order fusion channel.

[0157] As a possible implementation, the fusion unit 530 is configured to:

[0158] Perform defect fusion on a plurality of target defect data belonging to different target channels in the first fusion channel pair to obtain first fused defect data of a first-order fusion channel;

[0159] Defect fusion is performed on the first fused defect data belonging to different first-order fusion channels in the third fusion channel pair to obtain third fused defect data of the third-order fusion channel.

[0160] As a possible implementation, the optical characteristics include channel sensitivity, channel width, and light source type.

[0161] As a possible implementation, the multiple target defect data include defect type data, defect attribute data and defect location data; the fused defect data include type fusion data, attribute fusion data and location fusion data;

[0162] The fusion unit 530 is configured to:

[0163] Performing type fusion on multiple defect type data belonging to different target channels in the fusion channel pair to obtain type fusion data of the fusion channel pair;

[0164] Perform attribute fusion on multiple defect attribute data belonging to different target channels in the fusion channel pair to obtain attribute fusion data of the fusion channel pair;

[0165] Position fusion is performed on multiple defect position data belonging to different target channels in the fused channel pair to obtain position fusion data of the fused channel pair.

[0166] As a possible implementation, the fusion unit 530 is configured to:

[0167] If multiple defect type data belonging to different target channels in the fused channel pair are of the same defect type, the defect types of the type fused data of the fused channel pair and the multiple defect type data of the target channels are the same;

[0168] If multiple defect type data belonging to different target channels in the fused channel pair are different defect types, type fusion is performed on the multiple defect type data of different defect types according to a preset type fusion condition to obtain type fusion data of the fused channel pair.

[0169] As a possible implementation, the fusion unit 530 is configured to:

[0170] Get the coordinate weight values ​​of different target channels in the fusion channel pair;

[0171] The position fusion data of multiple defect positions are performed according to the preset coordinate fusion conditions and coordinate weight values ​​to obtain the position fusion data of the fusion channel pair.

[0172] As a possible implementation method, the preset coordinate fusion condition is to determine the defect location data with the largest coordinate weight value;

[0173] The fusion unit 530 is configured to:

[0174] The defect position data with the largest coordinate weight value is determined from the coordinate weight values ​​of different target channels as the position fusion data of the fusion channel pair.

[0175] As a possible implementation, the fusion unit 530 is configured to:

[0176] determining a fusion classification algorithm for the fusion channel pair according to the optical characteristics of the target channel included in the fusion channel pair;

[0177] According to the type defect data and the fusion classification algorithm of the fusion channel pair, attribute fusion is performed on multiple attribute defect data belonging to different target channels in the fusion channel pair to obtain attribute fusion data of the fusion channel pair.

[0178] As a possible implementation, the plurality of target defect data includes defect location data;

[0179] The fusion unit 530 is configured to:

[0180] Clustering the defect location data belonging to different target channels in the fused channel pair to obtain multiple clusters belonging to different target channels;

[0181] Matching defect location data included in a plurality of clusters respectively belonging to different target channels to obtain matched defect data respectively belonging to the different target channels;

[0182] Defect fusion is performed on the matching defect data belonging to different target channels to obtain fused defect data of the fused channel pair.

[0183] Based on the defect detection method provided in the above embodiment, the present application embodiment also provides a defect detection device, referring to Figure 6 FIG. 6 is a schematic diagram of a defect detection device according to an embodiment of the present application. The defect detection device 600 according to an embodiment of the present application includes:

[0184] An acquisition unit 610 is configured to acquire fusion defect data of a fusion channel pair;

[0185] The determining unit 620 is configured to determine, based on the fused defect data, a defect type, a defect attribute, and a defect location of a target defect corresponding to the fused defect data.

[0186] Based on the defect data fusion method provided in the above embodiment, the embodiment of the present application further provides a defect data fusion device, which includes:

[0187] The processor and the memory may be one or more processors. In some embodiments of the present application, the processor and the memory may be connected via a bus or other means.

[0188] The memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of the memory may also include NVRAM. The memory stores an operating system and operating instructions, executable modules, or data structures, or subsets or extended sets thereof. The operating instructions may include various operating instructions for implementing various operations. The operating system may include various system programs for implementing various basic services and processing hardware-based tasks.

[0189] The processor controls the operation of the terminal device and may also be referred to as a CPU.

[0190] The methods disclosed in the above embodiments of the present application can be applied to or implemented by a processor. The processor can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in the processor or by software instructions. The above processor can be a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. The methods, steps, and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present application can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0191] An embodiment of the present application also provides a computer-readable storage medium for storing program code, which is used to execute any implementation of the methods of the aforementioned embodiments.

[0192] In the context of the present application, a machine-readable medium can be a tangible medium that can contain or store a program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0193] It should be noted that the computer-readable medium mentioned above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0194] When introducing elements of various embodiments of the present application, the articles "a," "an," "the," and "said" are intended to mean that there are one or more elements. The words "comprising," "including," and "having" are inclusive and mean that there may be additional elements other than the listed elements.

[0195] It should be noted that those skilled in the art will appreciate that all or part of the processes in the above method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0196] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described above is merely illustrative, and the units and modules described as separate components may or may not be physically separated. In addition, some or all of the units and modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0197] The above is only a preferred embodiment of the present application. Although the present application has been disclosed as a preferred embodiment, it is not intended to limit the present application. Any technician familiar with the art can use the above-disclosed methods and technical contents to make many possible changes and modifications to the technical solution of the present application without departing from the scope of the technical solution of the present application, or modify it into an equivalent embodiment with equivalent changes. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application are still within the scope of protection of the technical solution of the present application.

Claims

1. A defect data fusion method, characterized in that: The method comprises: Acquire a plurality of target defect data of each target channel among a plurality of target channels and an optical characteristic of each target channel; determining at least one fusion channel pair with matching characteristics from the plurality of target channels according to the optical characteristics of each target channel; Defect fusion is performed on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair.

2. The method according to claim 1, characterized in that The optical characteristics include a plurality of optical dimensions, the plurality of optical dimensions including a first optical dimension and a second optical dimension, the defect variability in the first optical dimension being less than the defect variability in the second optical dimension; Determining at least one fusion channel pair with matching characteristics from the multiple target channels according to the optical characteristics of each target channel includes: determining, based on the first optical dimension, a first fused channel pair from the plurality of target channels that matches at least one characteristic of the first optical dimension; fusing the first fusion channel pair that performs characteristic matching on the first optical dimension to obtain a first-order fusion channel; determining, based on the second optical dimension, from the at least one first-order fused channel, a second fused channel pair having at least one matching characteristic in the second optical dimension; The second fusion channel pair that performs characteristic matching on the second optical dimension is fused to obtain a second-order fusion channel.

3. The method according to claim 2, characterized in that The plurality of optical dimensions further includes a third optical dimension; the method further comprising: Based on the third optical dimension, determining at least one third fused channel pair with characteristic matching in the third optical dimension from at least one second-order fused channel in the second optical dimension; fusing the third fused channel pair with characteristic matching in the third optical dimension to obtain a third-order fused channel; or, Based on the third optical dimension, at least one third fusion channel pair with characteristic matching in the third optical dimension is determined from the at least one first-order fusion channel; and the third fusion channel pair with characteristic matching in the third optical dimension is fused to obtain a third-order fusion channel.

4. The method according to claim 3, characterized in that The step of performing defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair includes: Performing defect fusion on a plurality of target defect data belonging to different target channels in the first fusion channel pair to obtain first fused defect data of the first-order fusion channel; performing defect fusion on first fused defect data belonging to different first-order fusion channels in the second fusion channel pair to obtain second fused defect data of the second-order fusion channel; Defect fusion is performed on the second fused defect data belonging to different second-order fusion channels in the third fusion channel pair to obtain third fused defect data of the third-order fusion channel.

5. The method according to claim 3, characterized in that The step of performing defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair includes: Performing defect fusion on a plurality of target defect data belonging to different target channels in the first fusion channel pair to obtain first fused defect data of the first-order fusion channel; Defect fusion is performed on the first fused defect data belonging to different first-order fusion channels in the third fusion channel pair to obtain third fused defect data of the third-order fusion channel.

6. The method according to claim 3, characterized in that The optical characteristics include channel sensitivity, channel width and light source type.

7. The method according to claim 1, characterized in that The plurality of target defect data include defect type data, defect attribute data and defect location data; the fused defect data include type fusion data, attribute fusion data and location fusion data; The step of performing defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair includes: Performing type fusion on a plurality of defect type data belonging to different target channels in the fusion channel pair to obtain type fusion data of the fusion channel pair; Performing attribute fusion on a plurality of defect attribute data belonging to different target channels in the fusion channel pair to obtain attribute fusion data of the fusion channel pair; Position fusion is performed on a plurality of defect position data belonging to different target channels in the fused channel pair to obtain position fusion data of the fused channel pair.

8. The method according to claim 7, characterized in that The performing type fusion on a plurality of defect type data belonging to different target channels in the fusion channel pair to obtain type fusion data of the fusion channel pair includes: If the multiple defect type data belonging to different target channels in the fused channel pair are of the same defect type, the defect type of the type fused data of the fused channel pair and the multiple defect type data of the target channel are the same; If the multiple defect type data belonging to different target channels in the fusion channel pair are different defect types, type fusion is performed on the multiple defect type data of different defect types according to a preset type fusion condition to obtain type fusion data of the fusion channel pair.

9. The method according to claim 7, characterized in that The performing position fusion on a plurality of defect position data belonging to different target channels in the fused channel pair to obtain position fusion data of the fused channel pair includes: Obtaining coordinate weight values ​​of different target channels in the fused channel pair; Position fusion data of the plurality of defect positions is performed according to a preset coordinate fusion condition and the coordinate weight value to obtain position fusion data of the fusion channel pair.

10. The method according to claim 9, characterized in that The preset coordinate fusion condition is to determine the defect position data with the largest coordinate weight value; The performing position fusion on the plurality of defect position data according to the preset coordinate fusion condition and the coordinate weight value to obtain the position fusion data of the fusion channel pair includes: Determine the defect position data with the largest coordinate weight value from the coordinate weight values ​​of the different target channels as the position fusion data of the fusion channel pair.

11. The method according to claim 7, characterized in that The step of fusing attributes of a plurality of defect attribute data belonging to different target channels in the fused channel pair to obtain attribute fusion data of the fused channel pair includes: Determining a fusion classification algorithm for the fusion channel pair according to optical characteristics of a target channel included in the fusion channel pair; Attribute fusion is performed on a plurality of attribute defect data belonging to different target channels in the fusion channel pair according to the type defect data and the fusion classification algorithm of the fusion channel pair to obtain attribute fusion data of the fusion channel pair.

12. The method according to any one of claims 1 to 11, characterized in that The plurality of target defect data includes defect location data; The step of performing defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair includes: Clustering the defect position data belonging to different target channels in the fused channel pair to obtain a plurality of clusters belonging to different target channels; Matching the defect location data included in the plurality of clusters respectively belonging to different target channels to obtain matched defect data respectively belonging to the different target channels; Defect fusion is performed on the matching defect data belonging to different target channels to obtain fused defect data of the fused channel pair.

13. A defect detection method, characterized in that: The defect data fusion method according to any one of claims 1 to 12, wherein the defect detection method comprises: Acquiring fusion defect data of the fusion channel pair; The defect type, defect attribute, and defect location of a target defect corresponding to the fused defect data are determined according to the fused defect data.

14. A defect data fusion device, characterized in that: include: an acquiring unit, configured to acquire a plurality of target defect data of each target channel among a plurality of target channels and an optical characteristic of each target channel; a determining unit, configured to determine at least one fusion channel pair with matching characteristics from the multiple target channels according to the optical characteristics of each target channel; The fusion unit is configured to perform defect fusion on a plurality of target defect data belonging to different target channels in the fusion channel pair to obtain fused defect data of the fusion channel pair.

15. A defect data fusion device, characterized in that: The device includes: a processor and a memory; The memory is used to store instructions; The processor is configured to execute the instructions in the memory to perform the method according to any one of claims 1 to 12.

16. A computer storage medium, characterized in that The computer storage medium is used to store a computer program, and when the computer program is run on a computer device, the computer device is enabled to execute the method according to any one of claims 1 to 12.

Images