Hidden danger handling method, device, electronic device and computer readable medium
Through identity verification, the information to be investigated is obtained, and the logical tree is identified and generated for hidden danger processing is solved, which solves the problem of the number of hidden danger inspectors and the area imbalance and improves the efficiency of hidden danger investigation.
Patent Information
- Application Number
- CN202410546610.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-06
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-05-06
AI Technical Summary
Among the existing hidden danger detection methods, the number of hidden danger detection personnel and the area is unbalanced, resulting in low efficiency in hidden danger detection, especially when safety hazards are found, which requires a long time to deal with, resulting in reduced inspection efficiency.
Through identity verification, obtain the set of hidden danger investigation information for the target place, conduct hidden danger identification to generate warning information, and generate hidden danger investigation logic tree based on the warning information, and automatically handle hidden dangers, so as to improve the self-inspection and self-correction efficiency of self-inspection objects for hidden dangers in the target place.
The efficiency of hidden danger detection has been improved. Through automated hidden danger identification and processing procedures, manual processing time has been reduced, and the efficiency and efficiency of hidden danger detection has been improved.
Smart Images

Figure CN118537186B_ABST
Abstract
Description
Technical Field
[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to a hidden danger handling method, device, electronic device, and computer-readable medium. Background Art
[0002] In recent years, safety accidents have frequently occurred in production and operation units, resulting in significant casualties and property losses. The cause of safety accidents is often due to improper handling of accident hazards. How to effectively detect hidden dangers has become an urgent problem to be solved. At present, when dealing with hidden dangers: the commonly used method is to have hidden danger detection personnel detect hidden dangers one by one.
[0003] However, when the above method is adopted, the following technical problems often occur:
[0004] There is a serious imbalance in the ratio between the number of hidden danger inspection personnel and the hidden danger inspection areas. Especially when safety hazards are discovered, the efficiency of hidden danger inspection is greatly reduced because it often takes a certain amount of time to deal with the hazards.
[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the invention
[0006] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.
[0007] Some embodiments of the present disclosure propose hidden danger handling methods, devices, electronic devices and computer-readable media to solve one or more of the technical problems mentioned in the above background technology section.
[0008] In a first aspect, some embodiments of the present disclosure provide a method for handling hidden dangers, the method comprising: performing identity authentication on a self-inspection object; in response to successful identity authentication, obtaining a set of information on hidden dangers to be checked corresponding to a target place, wherein the target place is a place associated with the self-inspection object and to be checked for hidden dangers, and the set of information on hidden dangers to be checked is information entered by the self-inspection object and corresponding to an area to be checked for hidden dangers; performing hidden danger identification on each information on hidden dangers to be checked in the set of information on hidden dangers to be checked to generate hidden danger warning information, wherein the hidden danger warning information comprises: a hidden danger type identifier, a hidden danger risk level and hidden danger rectification suggestion information, and the hidden danger type identifier indicates whether there are hidden dangers in the hidden danger inspection area corresponding to the information on hidden dangers to be checked; generating a hidden danger inspection logic tree based on the set of information on hidden dangers to be checked and the hidden danger warning information corresponding to the information on hidden dangers to be checked; and handling hidden dangers for the target place based on the hidden danger inspection logic tree.
[0009] In a second aspect, some embodiments of the present disclosure provide a hidden danger handling device, the device comprising: an identity authentication unit, configured to authenticate the identity of a self-inspection object; an acquisition unit, configured to acquire a set of information to be checked for hidden dangers corresponding to a target place in response to successful identity authentication, wherein the target place is a place associated with the self-inspection object to be checked for hidden dangers, and the set of information to be checked for hidden dangers is information entered by the self-inspection object and corresponding to an area to be checked for hidden dangers; a hidden danger identification unit, configured to identify hidden dangers for each hidden danger check in the set of information to be checked for hidden dangers, so as to generate hidden danger warning information, wherein the hidden danger warning information comprises: a hidden danger type identifier, a hidden danger risk level and hidden danger rectification suggestion information, and the hidden danger type identifier indicates whether there are hidden dangers in the hidden danger inspection area corresponding to the hidden danger check information; a generation unit, configured to generate a hidden danger inspection logic tree according to the set of information to be checked for hidden dangers and the hidden danger warning information corresponding to the hidden danger check information; a hidden danger handling unit, configured to handle hidden dangers for the target place according to the hidden danger inspection logic tree.
[0010] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the above-mentioned first aspect.
[0011] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner of the above-mentioned first aspect is implemented.
[0012] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the hidden danger handling method of some embodiments of the present disclosure, the efficiency of hidden danger investigation is improved. Specifically, the reason for the low efficiency of hidden danger investigation is that the number of hidden danger investigation personnel and hidden danger investigation areas is seriously unbalanced, especially when safety hazards are found, because hidden danger handling often takes a certain amount of time, which leads to a significant reduction in the efficiency of hidden danger investigation. In practice, hidden danger investigation usually requires professional personnel to conduct hidden danger investigation, for example, joint hidden danger investigation is conducted by personnel of industry regulatory units (fire, emergency bureau, etc.), safety officers of enterprises and institutions, etc. However, the number of the above-mentioned personnel is very different from the number of areas that need to be investigated for hidden dangers. Based on this, the hidden danger handling method of some embodiments of the present disclosure, first, the identity verification of the self-inspection object is performed. In practice, during the hidden danger investigation process, it may be necessary to enter the hidden danger investigation area. Therefore, the investigators should have the authority to enter the hidden danger investigation area, and thus, the identity verification of the self-inspection object is required. Secondly, in response to the identity verification being passed, the target place is obtained from the information set to be checked for hidden dangers, wherein the target place is a place to be checked for hidden dangers associated with the self-inspection object, and the information set to be checked for hidden dangers is the information corresponding to the area to be checked for hidden dangers entered by the self-inspection object. In this way, the information that needs to be checked for hidden dangers is entered by the self-inspection object. Then, each hidden danger check information in the above hidden danger check information set is identified for hidden dangers to generate hidden danger warning information, wherein the hidden danger warning information includes: hidden danger type identification, hidden danger risk level and hidden danger rectification suggestion information, and the hidden danger type identification indicates whether there are hidden dangers in the hidden danger check area corresponding to the hidden danger check information. In this way, the hidden danger type, hidden danger risk and hidden danger rectification suggestion are automatically identified. Then, according to the above hidden danger check information set and the hidden danger warning information corresponding to the hidden danger check information, a hidden danger check logic tree is generated. Finally, according to the hidden danger check logic tree, the target place is handled for hidden dangers. In this way, the self-inspection object realizes self-inspection and self-correction of hidden dangers in the target place, and improves the efficiency of hidden danger checking. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.
[0014] Figure 1 is a flow chart of some embodiments of the hidden danger handling method according to the present disclosure;
[0015] Figure 2 is a schematic structural diagram of some embodiments of the hidden danger handling device according to the present disclosure;
[0016] Figure 3 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION
[0017] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.
[0018] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.
[0019] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.
[0020] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".
[0021] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.
[0022] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.
[0023] refer to Figure 1 , shows a process 100 of some embodiments of the hidden danger handling method according to the present disclosure. The hidden danger handling method comprises the following steps:
[0024] Step 101, identity verification is performed on the self-inspection object.
[0025] In some embodiments, the execution subject (e.g., a computing device) of the hidden danger handling method can authenticate the identity of the self-inspection object. Among them, the self-inspection object can be an object associated with the target place. For example, the self-inspection object can be the owner or operator of the target place. The target place can be a place associated with the above-mentioned self-inspection object to be checked for hidden dangers. The self-inspection object can inspect and handle hidden dangers in the target place by self-inspection. In practice, the above-mentioned execution subject can authenticate the identity of the self-inspection object by means of a text message verification code through the phone number corresponding to the self-inspection object.
[0026] It should be noted that the above-mentioned computing device can be hardware or software. When the computing device is hardware, it can be implemented as a single terminal device. When the computing device is embodied as software, it can be installed in the hardware devices listed above. It can be implemented as a single software or software module. No specific limitation is made here.
[0027] In some optional implementations of some embodiments, the execution subject authenticates the identity of the self-check object, which may include the following steps:
[0028] The first step is to determine the calibration type.
[0029] Among them, the verification types include: biometric verification type and identity information verification type. The biometric verification type indicates that identity verification is performed through the biometric corresponding to the self-inspection object. The identity information verification type indicates that identity verification is performed through the identity information corresponding to the self-inspection object. In practice, the verification type can be selected by the self-inspection object.
[0030] In the second step, in response to determining that the verification type is a biometric verification type, the biometrics of the self-examination object are obtained.
[0031] Among them, the biometric features include: a facial image set. The facial images in the above facial image set are collected at different angles. In practice, the above self-examination subject can collect the facial image set through the front camera included in the mobile terminal (for example, a smart phone) and send it to the above execution subject. Specifically, when collecting facial images, the mobile terminal can prompt the self-examination subject to turn his head by displaying a prompt message such as "Please turn your head" to collect facial images at different angles to obtain the above facial image set.
[0032] The third step is to extract key points from each facial image in the facial image set to generate a key point map and obtain a key point map set.
[0033] In practice, the execution subject may extract key points from each facial image in the facial image set by means of Hough transform to generate a key point map and obtain a key point map set.
[0034] Optionally, performing key point extraction on each facial image in the facial image set to generate a key point map may include the following steps:
[0035] The first sub-step is to perform grayscale processing on the facial image to generate a grayscale processed image.
[0036] In practice, the execution subject may use a weighted average method to grayscale the facial image to generate a grayscale processed image. Specifically, the execution subject may use the average of the color values of each pixel in the facial image in the red channel, the green channel, and the blue channel as the color value corresponding to the pixel to obtain the grayscale processed image.
[0037] In the second sub-step, in response to the image size of the grayscale processed image being inconsistent with a preset size, the grayscale processed image is cropped to obtain a cropped image.
[0038] Wherein, the image size of the cropped image is the preset size.
[0039] The third sub-step is to generate a facial key point set based on the cropped image and the facial key point extraction model.
[0040] In practice, the facial key point extraction model may be a model for extracting key points contained in a facial image. The facial key point extraction model takes a facial image as input and outputs the identified key points. In practice, the facial key point extraction model may adopt a Resnet-50 model.
[0041] In the fourth sub-step, the facial key points are grouped according to the relative positions of the facial key points in the above facial key point set to obtain a facial key point group set.
[0042] The facial key point group includes three facial key points that are relatively close to each other. In practice, the execution subject can group the facial key points according to the relative positions of the facial key points in the facial key point set by using a K-nearest neighbor algorithm to obtain a facial key point group set. Specifically, the value of K is 3.
[0043] The fifth sub-step is to determine the regional midpoint of each facial key point group in the above facial key point group set to obtain a regional midpoint set.
[0044] In practice, the execution subject may determine the midpoint of a triangular area enclosed by the facial key points in the facial key point group.
[0045] The sixth sub-step is to grid the cropped image to generate a gridded image, wherein the gridded image includes at least one grid block.
[0046] As an example, the image size of the cropped image may be M×M, where M×M is the image resolution of the cropped image. Then the size of the grid block may be 1×1. Specifically, the execution subject may grid the cropped image in a horizontal scanning manner to generate a gridded image. Therefore, the gridded image may be represented in the form of a matrix, and the matrix size is M×M.
[0047] The seventh sub-step is to determine the grid blocks corresponding to the regional midpoints in the regional midpoint set in the gridded image, and the offsets from the centers of the corresponding grid blocks, so as to generate the key point map.
[0048] In practice, first, the above-mentioned execution subject can project the regional midpoint into the gridded image to determine the grid block corresponding to the regional midpoint. Then, determine the lateral offset and longitudinal offset between the regional midpoint and the corresponding grid block to obtain the offset. Since the gridded image can be represented in matrix form, the matrix value in the matrix corresponds to the grid block. Specifically, the matrix value can be [P, O1, O2], where P represents whether there is a regional midpoint in the grid block corresponding to the matrix value. When it exists, P takes the value of 1, and when it does not exist, P takes the value of 0, and O1 also takes the value of 0 and O2 also takes the value of 0. O1 represents the lateral offset. O2 represents the longitudinal offset. Therefore, the key point map can be represented by a matrix with a matrix dimension of M×M×3.
[0049] In the fourth step, feature map matching is performed based on the above key point map set and the facial feature maps in the pre-built facial feature map library to generate a verification result.
[0050] Among them, the verification results include: identity authentication passed and identity authentication failed. In practice, for each key point graph in the key point graph set, the above-mentioned execution entity can convert the key point graph into a one-dimensional feature vector. Among them, the vector value in the one-dimensional feature vector is the key point coordinate in the key point graph. Then, the vector similarity between the one-dimensional feature vector and the one-dimensional vector corresponding to the facial feature graph in the facial feature graph library is calculated. Then, when the similarities of each vector in the obtained vector similarity set are greater than the similarity threshold or the average value of each vector similarity is greater than the similarity threshold, a verification result representing that the identity authentication passed is generated, otherwise, a verification result representing that the identity authentication failed is generated.
[0051] In a fifth step, in response to determining that the verification type is an identity information verification type, the identity information input by the self-inspection subject is obtained.
[0052] In practice, the identity information may include: the name of the self-inspection subject, the telephone number of the self-inspection subject, the ID number of the self-inspection subject, and the name of the self-inspection location. For example, first, the self-inspection subject may input the identity information through a mobile terminal. Then, the mobile terminal uses the MurMurHash algorithm to hash the identity information to obtain a hash string. Next, the mobile terminal encrypts the hash string and sends it to the execution subject as identity information.
[0053] The sixth step is to decrypt the above identity information to obtain the decrypted identity information.
[0054] The decrypted identity information is a hash string that has been hashed. In practice, the execution subject may use a decryption algorithm corresponding to the information encryption of the hash string by the mobile terminal to decrypt the identity information and obtain the decrypted identity information.
[0055] The seventh step is to perform identity authentication based on the decrypted identity information and the pre-built identity information database to generate a verification result.
[0056] The identity information of different self-check objects is pre-entered in the identity information database. In practice, the above-mentioned execution subject can compare the decrypted identity information with the identity information in the identity information database by string comparison to generate a verification result. For example, when the identity information that is the same as the decrypted identity information exists in the identity information database, a verification result indicating that the identity authentication is passed is generated; when the identity information that is the same as the decrypted identity information does not exist in the identity information database, a verification result indicating that the identity authentication is not passed is generated.
[0057] Step 102, in response to identity verification being passed, obtaining a set of information on hidden dangers to be checked corresponding to the target location.
[0058] In some embodiments, the above-mentioned execution subject can obtain a set of information on hidden dangers to be checked corresponding to the target place in response to identity verification. The target place is a place associated with the above-mentioned self-inspection object and to be checked for hidden dangers. The set of information on hidden dangers to be checked is information corresponding to the area to be checked for hidden dangers, entered by the above-mentioned self-inspection object. Specifically, the information on hidden dangers to be checked can be regional description information for the hidden danger inspection area. In practice, the above-mentioned execution subject can obtain the description information of the area to be checked for hidden dangers collected by the self-inspection object through a mobile terminal for the hidden danger inspection area, and send it to the above-mentioned execution subject. For example, the above-mentioned self-inspection object can enter the description information of the area to be checked for hidden dangers by filling out a questionnaire.
[0059] Optionally, the information to be checked for hidden dangers in the information set to be checked for hidden dangers includes: a description information set of the area to be checked for hidden dangers. The information types of the description information of the area to be checked for hidden dangers include: image type, text type and video type. For example, when the information type of the description information of the area to be checked for hidden dangers is an image type, it may be an image obtained by collection for the hidden danger checking area. For another example, when the information type of the description information of the area to be checked for hidden dangers is a text type, it may be a text description of the hidden danger checking area input by the self-inspection object. For another example, when the information type of the description information of the area to be checked for hidden dangers is a video type, it may be a video obtained by collection for the hidden danger checking area.
[0060] Step 103, identifying hidden dangers for each hidden danger inspection information in the hidden danger inspection information set to generate hidden danger warning information.
[0061] In some embodiments, the execution subject may identify hidden dangers for each hidden danger inspection information in the hidden danger inspection information set to generate hidden danger warning information. The hidden danger warning information represents the hidden danger warning description for the hidden danger inspection area. The hidden danger warning information includes: hidden danger type identification, hidden danger risk level and hidden danger rectification suggestion information. The hidden danger type identification represents whether there is a hidden danger in the hidden danger inspection area corresponding to the hidden danger inspection information.
[0062] In some optional implementations of some embodiments, the execution subject identifies hidden dangers for each hidden danger inspection information in the hidden danger inspection information set to generate hidden danger warning information, which may include the following steps:
[0063] The first step is to activate the multimodal feature extraction block in the hidden danger warning information generation model, which corresponds to the information type of the description information of the hidden danger inspection area included in the hidden danger inspection information, and obtain the activated hidden danger warning information generation model.
[0064] Among them, the hidden danger warning information generation model includes: 3 multimodal feature extraction blocks, 7 feature fusion blocks, 1 hidden danger type classifier, 1 hidden danger risk level classifier and hidden danger rectification suggestion information prediction head. The multimodal feature extraction blocks correspond one-to-one to the information type of the description information of the area to be checked for hidden dangers. The hidden danger type classifier is a binary classifier. Specifically, the three multimodal feature extraction blocks include: multimodal feature extraction block A, multimodal feature extraction block B and multimodal feature extraction block C. Among them, multimodal feature extraction block A, multimodal feature extraction block B and multimodal feature extraction block C are set in parallel. Multimodal feature extraction block A is used to extract features from the description information of the area to be checked for hidden dangers of image type. Multimodal feature extraction block B is used to extract features from the description information of the area to be checked for hidden dangers of text type. Multimodal feature extraction block C is used to extract features from the description information of the area to be checked for hidden dangers of video type. The multimodal feature extraction block A adopts the encoding idea of the BERT model. Specifically, first, the description information of the area to be inspected for hidden dangers of the multimodal feature extraction block A image type is divided into multiple image blocks. Then, the image block is masked. Next, the masked image block is feature encoded by the encoder. For example, the multimodal feature extraction block A can adopt the BEiT model. The multimodal feature extraction block B adopts the encoder structure in the BERT model. Since the video is a collection of images, the multimodal feature extraction block C can reuse the model structure of the multimodal feature extraction block A. The 7 feature fusion blocks may include: feature fusion block A, feature fusion block B, feature fusion block C, feature fusion block D, feature fusion block E, feature fusion block F and feature fusion block G. Among them, the output sizes of feature fusion block A, feature fusion block B, feature fusion block C, feature fusion block D, feature fusion block E, feature fusion block F and feature fusion block G are consistent. Specifically, the input of feature fusion block A is the output of multimodal feature extraction block A. The input of feature fusion block B is the output of multimodal feature extraction block B. The input of feature fusion block C is the output of multimodal feature extraction block C. The input of feature fusion block D is the output of multimodal feature extraction block A and the output of multimodal feature extraction block B. The input of feature fusion block E is the output of multimodal feature extraction block A and the output of multimodal feature extraction block C. The input of feature fusion block F is the output of multimodal feature extraction block B and the output of multimodal feature extraction block C. The input of feature fusion block G is the output of multimodal feature extraction block A, the output of multimodal feature extraction block B and the output of multimodal feature extraction block C. In practice, feature fusion block A, feature fusion block B, feature fusion block C, feature fusion block D, feature fusion block E, feature fusion block F and feature fusion block G all adopt the decoder structure as the basic structure. In addition, considering that the inputs of feature fusion block D, feature fusion block E, feature fusion block F and feature fusion block G are multiple and the output sizes of the 7 feature fusion blocks are required to be consistent.Therefore, feature fusion block D, feature fusion block E, feature fusion block F and feature fusion block G all further include a feature overlay layer and a feature mapping layer. Among them, the feature overlay layer is used to overlay multiple features. For example, the feature overlay layer can use the add method to overlay features. The feature mapping layer is used to control the output size of the feature fusion block to be consistent. For example, the feature mapping layer can unify the feature dimension by downsampling. The hidden danger type classifier is a binary classifier, and the classification result can be [0, 1] or [1, 0]. [0, 1] represents the hidden danger type identification as "1", that is, there is a safety hazard, and [0, 1] represents the hidden danger type identification as "0", that is, there is no safety hazard. The hidden danger risk level classifier is a multi-classifier, and the classification result is determined according to the number of preset hidden danger risk levels. For example, the hidden danger risk levels can include: level 1, level 2, level 3, level 4 and level 5, and the vector dimension of the classification result output by the hidden danger risk level classifier is 1×5. The hidden danger rectification suggestion information prediction head can use the GPT model. Considering that the three multimodal feature extraction blocks have performed multimodal feature extraction, in order to avoid repeated feature extraction, reduce the model structure and reduce the model complexity, the hidden danger rectification suggestion information prediction head can adopt the Text Predicition part in the GPT model.
[0065] The second step is to generate hidden danger warning information corresponding to the hidden danger inspection information according to the activated hidden danger warning information generation model and the description information set of the hidden danger inspection area included in the hidden danger inspection information.
[0066] Step 104: Generate a hidden danger checking logic tree according to the hidden danger checking information set and the hidden danger warning information corresponding to the hidden danger checking information.
[0067] In some embodiments, the execution subject may generate a hidden danger troubleshooting logic tree based on the hidden danger troubleshooting information set and the hidden danger warning information corresponding to the hidden danger troubleshooting information. The hidden danger troubleshooting logic tree represents a tree structure used to guide the self-inspection object to conduct hidden danger self-inspection and treatment on the target site. For example, the execution subject may use the hidden danger troubleshooting information and the corresponding hidden danger warning information as tree nodes to form a tree structure as a hidden danger troubleshooting logic tree.
[0068] Optionally, the hidden dangers to be checked information in the hidden dangers to be checked information set further includes: the location of the area to be checked for hidden dangers. The location of the area to be checked for hidden dangers represents the location coordinates of the area corresponding to the hidden dangers to be checked information in the target production site.
[0069] In some optional implementations of some embodiments, the execution subject generates a hidden danger troubleshooting logic tree according to the hidden danger troubleshooting information set and the hidden danger warning information corresponding to the hidden danger troubleshooting information, which may include the following steps:
[0070] The first step is to cluster the above-mentioned information set to be checked for hidden dangers according to the location of the area to be checked for hidden dangers included in the information to be checked for hidden dangers in the above-mentioned information set to be checked for hidden dangers, so as to obtain a sequence of information groups to be checked for hidden dangers.
[0071] Among them, the hidden dangers to be checked information in the hidden dangers to be checked information group corresponds to the nearby hidden dangers to be checked area location. For example, the above-mentioned execution subject can cluster the above-mentioned hidden dangers to be checked information set according to the hidden dangers to be checked area location included in the hidden dangers to be checked information in the above-mentioned hidden dangers to be checked information set through the K-means clustering algorithm to obtain the hidden dangers to be checked information group sequence.
[0072] The second step is to construct an initial hidden danger inspection logic tree based on the target hidden danger inspection information group.
[0073] The initial hidden danger troubleshooting logic tree includes: a root node and a sub-node set. The target hidden danger troubleshooting information group is the hidden danger troubleshooting information group located at the median position in the sequence of the hidden danger troubleshooting information groups. The number of sub-nodes in the sub-node set + 1 is the same as the number of target hidden danger troubleshooting information in the target hidden danger troubleshooting information group.
[0074] As an example, the sequence of information groups to be checked for hidden dangers includes: information group A to be checked for hidden dangers, information group B to be checked for hidden dangers, and information group C to be checked for hidden dangers. Information group A to be checked for hidden dangers includes: information A1 to be checked for hidden dangers, information A2 to be checked for hidden dangers, and information A3 to be checked for hidden dangers. Information group B to be checked for hidden dangers includes: information B1 to be checked for hidden dangers, information B2 to be checked for hidden dangers, information B3 to be checked for hidden dangers, and information B4 to be checked for hidden dangers. Information group C to be checked for hidden dangers includes: information C1 to be checked for hidden dangers, and information C2 to be checked for hidden dangers. Therefore, the target information group to be checked for hidden dangers is information group B to be checked for hidden dangers. The root node included in the initial hidden danger checking logic tree corresponds to any information to be checked for hidden dangers in information group B to be checked for hidden dangers, for example, corresponds to information B1 to be checked for hidden dangers. Then the set of child nodes included in the initial hidden danger checking logic tree corresponds to information B2 to be checked for hidden dangers, information B3 to be checked for hidden dangers, and information B4 to be checked for hidden dangers.
[0075] The third step is to update the initial hidden danger troubleshooting logic tree according to the hidden danger troubleshooting information groups in the hidden danger troubleshooting information group sequence except the target hidden danger troubleshooting information group to be checked, so as to generate an initial hidden danger troubleshooting logic tree after tree update.
[0076] In practice, the execution subject can determine the sub-nodes linked to the information groups to be checked for hidden dangers in the sequence of the information groups to be checked for hidden dangers, excluding the target information groups to be checked for hidden dangers, in the sub-node set included in the initial hidden danger checking logic tree by using the remainder formula. The remainder formula is: S=A mod B, wherein A represents the position of the information groups to be checked for hidden dangers in the sequence of the information groups to be checked for hidden dangers, excluding the target information groups to be checked for hidden dangers, in the sequence of the information groups to be checked for hidden dangers. B represents the number of sub-nodes in the sub-node set included in the initial hidden danger checking logic tree. S represents the position of the sub-nodes linked to the information groups to be checked for hidden dangers, excluding the target information groups to be checked for hidden dangers, in the sub-node set included in the initial hidden danger checking logic tree.
[0077] As an example, information group A to be checked for hidden dangers is linked to the sub-node corresponding to information B2 to be checked for hidden dangers. Information group C to be checked for hidden dangers is linked to the sub-node corresponding to information B3 to be checked for hidden dangers. Specifically, information group A to be checked for hidden dangers includes information A1 to be checked for hidden dangers, information A2 to be checked for hidden dangers, and information A3 to be checked for hidden dangers, which are sub-nodes of the sub-node corresponding to information B2 to be checked for hidden dangers. Information group C to be checked for hidden dangers includes information C1 to be checked for hidden dangers and information C2 to be checked for hidden dangers, which are sub-nodes of the sub-node corresponding to information B3 to be checked for hidden dangers.
[0078] The fourth step is to adjust the tree structure of the initial hidden danger investigation logic tree after the tree update according to the hidden danger risk level included in the hidden danger warning information corresponding to the hidden danger investigation information to be investigated, so as to obtain the hidden danger investigation logic tree.
[0079] As an example, the parent node of the child node corresponding to the hidden dangers to be checked B3 is the root node corresponding to the hidden dangers to be checked B1. The parent node of the child node corresponding to the hidden dangers to be checked A1 is the child node corresponding to the hidden dangers to be checked B3. When the hidden danger risk level included in the hidden danger warning information corresponding to the hidden dangers to be checked A1 is greater than the hidden danger risk level included in the hidden danger warning information corresponding to the hidden dangers to be checked B3, delete the edge between the child node corresponding to the hidden dangers to be checked B3 and the child node corresponding to the hidden dangers to be checked A1, and create an edge between the child node corresponding to the hidden dangers to be checked A1 and the root node corresponding to the hidden dangers to be checked B1, so as to adjust the tree structure.
[0080] Step 105, according to the hidden danger investigation logic tree, handle the hidden dangers at the target location.
[0081] In practice, the above-mentioned execution subject can extensively traverse the above-mentioned hidden danger investigation logic tree, and perform hidden danger self-inspection and processing according to the hidden danger investigation information and corresponding hidden danger warning information corresponding to the nodes in the hidden danger investigation logic tree.
[0082] Optionally, the above method further includes:
[0083] The first step is to prompt the above-mentioned self-inspection object to re-enter the hidden danger investigation information after the hidden danger is handled in response to the hidden danger being handled.
[0084] In practice, after the self-inspection subject has confirmed the hidden danger rectification through the mobile terminal, the above-mentioned execution entity can send a hidden danger inspection information re-entry reminder to the mobile terminal to remind the self-inspection subject to re-enter the hidden danger inspection information.
[0085] In the second step, in response to the completion of the input, hidden dangers are re-identified based on the hidden danger inspection information after the re-entry, so as to generate hidden danger warning information after the re-identification.
[0086] Specifically, the above-mentioned execution entity can use the hidden danger identification method in step 103 to identify hidden dangers, which will not be repeated here.
[0087] The third step is to update the hidden danger investigation logic tree in real time according to the hidden danger warning information after the above re-identification.
[0088] In practice, the above-mentioned execution entity may update the hidden danger warning information corresponding to the tree node corresponding to the re-identified hidden danger warning information in the hidden danger investigation logic tree, so as to realize real-time tree update of the hidden danger investigation logic tree.
[0089] The fourth step is to display the progress of hidden danger treatment in the above-mentioned target places in real time according to the above-mentioned hidden danger investigation logic tree.
[0090] In practice, the above-mentioned executor can use a large digital screen to display in real time the progress of handling hidden dangers corresponding to the nodes in the hidden danger investigation logic tree, so as to achieve the purpose of displaying the progress of hidden danger handling in the above-mentioned target place in real time.
[0091] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the hidden danger handling method of some embodiments of the present disclosure, the efficiency of hidden danger investigation is improved. Specifically, the reason for the low efficiency of hidden danger investigation is that the number of hidden danger investigation personnel and hidden danger investigation areas is seriously unbalanced, especially when safety hazards are found, because hidden danger handling often takes a certain amount of time, which leads to a significant reduction in the efficiency of hidden danger investigation. In practice, hidden danger investigation usually requires professional personnel to conduct hidden danger investigation, for example, joint hidden danger investigation is conducted by personnel of industry regulatory units (fire, emergency bureau, etc.), safety officers of enterprises and institutions, etc. However, the number of the above-mentioned personnel is very different from the number of areas that need to be investigated for hidden dangers. Based on this, the hidden danger handling method of some embodiments of the present disclosure, first, the identity verification of the self-inspection object is performed. In practice, during the hidden danger investigation process, it may be necessary to enter the hidden danger investigation area. Therefore, the investigators should have the authority to enter the hidden danger investigation area, and thus, the identity verification of the self-inspection object is required. Secondly, in response to the identity verification being passed, the target place is obtained from the information set to be checked for hidden dangers, wherein the target place is a place to be checked for hidden dangers associated with the self-inspection object, and the information set to be checked for hidden dangers is the information corresponding to the area to be checked for hidden dangers entered by the self-inspection object. In this way, the information that needs to be checked for hidden dangers is entered by the self-inspection object. Then, each hidden danger check information in the above hidden danger check information set is identified for hidden dangers to generate hidden danger warning information, wherein the hidden danger warning information includes: hidden danger type identification, hidden danger risk level and hidden danger rectification suggestion information, and the hidden danger type identification indicates whether there are hidden dangers in the hidden danger check area corresponding to the hidden danger check information. In this way, the hidden danger type, hidden danger risk and hidden danger rectification suggestion are automatically identified. Then, according to the above hidden danger check information set and the hidden danger warning information corresponding to the hidden danger check information, a hidden danger check logic tree is generated. Finally, according to the hidden danger check logic tree, the target place is handled for hidden dangers. In this way, the self-inspection object realizes self-inspection and self-correction of hidden dangers in the target place, and improves the efficiency of hidden danger checking.
[0092] Further references Figure 2 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a hidden danger handling device. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the hidden danger handling device can be specifically applied to various electronic devices.
[0093] like Figure 2As shown, the hidden danger handling device 200 of some embodiments includes: an identity authentication unit 201, an acquisition unit 202, a hidden danger identification unit 203, a generation unit 204 and a hidden danger handling unit 205. Among them, the identity authentication unit 201 is configured to perform identity authentication on the self-inspection object; the acquisition unit 202 is configured to obtain a set of information on hidden dangers to be checked corresponding to the target place in response to the identity authentication being passed, wherein the above-mentioned target place is a place associated with the above-mentioned self-inspection object and to be checked for hidden dangers, and the above-mentioned information set to be checked for hidden dangers is information entered by the above-mentioned self-inspection object and corresponding to the area to be checked for hidden dangers; the hidden danger identification unit 203 is configured to identify hidden dangers for each hidden danger to be checked in the above-mentioned information set to be checked for hidden dangers, so as to generate hidden danger warning information, wherein the above-mentioned hidden danger warning information includes: hidden danger type identification, hidden danger risk level and hidden danger rectification suggestion information, and the above-mentioned hidden danger type identification indicates whether there are hidden dangers in the hidden danger inspection area corresponding to the hidden danger inspection information to be checked; the generation unit 204 is configured to generate a hidden danger inspection logic tree according to the above-mentioned information set to be checked for hidden dangers and the hidden danger warning information corresponding to the hidden danger inspection information; the hidden danger processing unit 205 is configured to process hidden dangers for the above-mentioned target place according to the above-mentioned hidden danger inspection logic tree.
[0094] It is understandable that the various units recorded in the hidden danger handling device 200 are similar to those in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the hidden danger handling device 200 and the units contained therein, and will not be repeated here.
[0095] Reference below Figure 3 , which shows a structural schematic diagram of an electronic device (eg, a computing device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.
[0096] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory 302 or a program loaded from a storage device 308 into a random access memory 303. Various programs and data required for the operation of the electronic device 300 are also stored in the random access memory 303. The processing device 301, the read-only memory 302, and the random access memory 303 are connected to each other via a bus 304. An input / output interface 305 is also connected to the bus 304.
[0097] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.
[0098] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the read-only memory 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.
[0099] It should be noted that the computer-readable medium recorded in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may 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 may 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 some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit 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 of the above.
[0100] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (Hyper Text Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.
[0101] The computer-readable medium may be included in the electronic device; or it may exist independently without being installed in the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: performs identity authentication on the self-inspection object; in response to the identity authentication being passed, obtains a set of hidden danger inspection information corresponding to the target place, wherein the target place is a place associated with the self-inspection object and to be inspected for hidden dangers, and the set of hidden danger inspection information is information corresponding to the area to be inspected for hidden dangers entered by the self-inspection object; identifies hidden dangers for each hidden danger inspection information in the set of hidden danger inspection information to generate hidden danger warning information, wherein the hidden danger warning information includes: hidden danger type identification, hidden danger risk level and hidden danger rectification suggestion information, and the hidden danger type identification indicates whether there are hidden dangers in the hidden danger inspection area corresponding to the hidden danger inspection information; generates the hidden danger inspection logic tree according to the set of hidden danger inspection information and the hidden danger warning information corresponding to the hidden danger inspection information; and handles hidden dangers for the target place according to the hidden danger inspection logic tree.
[0102] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0103] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession 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 square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0104] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be arranged in a processor, for example, may be described as: a processor includes an identity authentication unit, an acquisition unit, a hidden danger identification unit, a generation unit and a hidden danger processing unit. Among them, the names of these units do not constitute a limitation on the units themselves under certain circumstances. For example, the hidden danger processing unit may also be described as "a unit that performs hidden danger processing on the above-mentioned target place according to the above-mentioned hidden danger investigation logic tree".
[0105] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.
[0106] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.
Claims
1. A method for handling hidden dangers, comprising: Authentication of the self-inspection object; In response to identity verification being passed, obtaining a set of information on hidden dangers to be checked corresponding to a target place, wherein the target place is a place associated with the self-inspection object and to be checked for hidden dangers, the set of information on hidden dangers to be checked is information corresponding to an area to be checked for hidden dangers entered by the self-inspection object, and the information on hidden dangers to be checked in the set of information on hidden dangers to be checked includes: a description information set of an area to be checked for hidden dangers and a location of an area to be checked for hidden dangers, and the information type of the description information of the area to be checked for hidden dangers includes: an image type, a text type, and a video type; Performing hidden danger identification on each hidden danger inspection information in the hidden danger inspection information set to generate hidden danger warning information, wherein the hidden danger warning information includes: hidden danger type identification, hidden danger risk level and hidden danger rectification suggestion information, and the hidden danger type identification indicates whether there is a hidden danger in the hidden danger inspection area corresponding to the hidden danger inspection information; Generate a hidden danger investigation logic tree according to the hidden danger investigation information set and the hidden danger warning information corresponding to the hidden danger investigation information; According to the hidden danger investigation logic tree, hidden danger treatment is performed on the target location, wherein: Performing hidden danger identification on each hidden danger inspection information in the hidden danger inspection information set to generate hidden danger warning information, including: Activate the multimodal feature extraction block in the hidden danger warning information generation model that corresponds to the information type of the description information of the area to be checked for hidden dangers included in the information to be checked for hidden dangers, and obtain the activated hidden danger warning information generation model, wherein the hidden danger warning information generation model includes: 3 multimodal feature extraction blocks, 7 feature fusion blocks, 1 hidden danger type classifier, 1 hidden danger risk level classifier and hidden danger rectification suggestion information prediction head, the multimodal feature extraction blocks correspond to the information type of the description information of the area to be checked for hidden dangers one by one, and the hidden danger type classifier is a binary classifier; According to the activated hidden danger warning information generation model and the hidden danger inspection area description information set included in the hidden danger inspection information, hidden danger warning information corresponding to the hidden danger inspection information is generated, wherein: The generating of a hidden danger checking logic tree according to the hidden danger checking information set and the hidden danger warning information corresponding to the hidden danger checking information includes: Clustering the set of information to be checked for hidden dangers according to the position of the area to be checked for hidden dangers included in the information to be checked for hidden dangers in the set of information to be checked for hidden dangers to obtain a sequence of information groups to be checked for hidden dangers; An initial hidden danger troubleshooting logic tree is constructed with the target hidden danger troubleshooting information group, wherein the initial hidden danger troubleshooting logic tree includes: a root node and a child node set, the target hidden danger troubleshooting information group is the hidden danger troubleshooting information group located at the median position in the hidden danger troubleshooting information group sequence, and the number of child nodes in the child node set + 1 is the same as the number of target hidden danger troubleshooting information in the target hidden danger troubleshooting information group; According to the information groups to be checked for hidden dangers in the sequence of information groups to be checked for hidden dangers, excluding the target information group to be checked for hidden dangers, the initial hidden danger checking logic tree is updated to generate an initial hidden danger checking logic tree after tree update; According to the hidden danger risk level included in the hidden danger warning information corresponding to the hidden danger inspection information, the tree structure of the initial hidden danger inspection logic tree after the tree update is adjusted to obtain the hidden danger inspection logic tree.
2. The method according to claim 1, wherein: The method further comprises: In response to the hidden danger being handled, prompting the self-inspection subject to re-enter the hidden danger troubleshooting information after the hidden danger is handled; In response to the completion of the input, the hidden dangers are re-identified according to the hidden danger investigation information after the re-input, so as to generate the hidden danger warning information after the re-identification; According to the re-identified hidden danger warning information, the hidden danger investigation logic tree is updated in real time.
3. The method according to claim 2, wherein: The method further comprises: According to the hidden danger investigation logic tree, the hidden danger handling progress of the target place is displayed in real time.
4. The method according to claim 3, wherein: The identity verification of the self-examination object includes: Determine a verification type, wherein the verification type includes: a biometric verification type and an identity information verification type; In response to determining that the verification type is a biometric verification type, obtaining a biometric of the self-examination object, wherein the biometric includes: a facial image set, each facial image in the facial image set is acquired at a different angle; Extracting key points from each facial image in the facial image set to generate a key point map, thereby obtaining a key point map set; Performing feature map matching according to the key point map set and the facial feature map in the pre-built facial feature map library to generate a verification result, wherein the verification result includes: identity authentication passed and identity authentication failed; In response to determining that the verification type is an identity information verification type, obtaining identity information input by the self-checking subject; Decrypting the identity information to obtain decrypted identity information, wherein the decrypted identity information is a hash character string that has been hashed; According to the decrypted identity information and the pre-built identity information database, identity verification is performed to generate a verification result.
5. The method according to claim 4, wherein: The step of extracting key points from each facial image in the facial image set to generate a key point map includes: Performing grayscale processing on the facial image to generate a grayscale processed image; In response to the image size of the grayscale processed image being inconsistent with a preset size, cropping the grayscale processed image to obtain a cropped image, wherein the image size of the cropped image is the preset size; Generate a facial key point set according to the cropped image and the facial key point extraction model; According to the relative positions of the facial key points in the facial key point set, the facial key points are grouped to obtain a facial key point group set, wherein the facial key point group includes three facial key points that are adjacent in relative position; Determine the regional midpoint of each facial key point group in the facial key point group set to obtain a regional midpoint set; Performing image gridding on the cropped image to generate a gridded image, wherein the gridded image includes: at least one grid block; The grid blocks corresponding to the regional points in the set of regional points in the gridded image and the offsets from the centers of the corresponding grid blocks are determined to generate the key point map.
6. A hidden danger handling device, comprising: an identity verification unit configured to verify the identity of the self-inspection object; an acquisition unit, configured to acquire, in response to identity verification being passed, a set of information to be checked for hidden dangers corresponding to a target place, wherein the target place is a place associated with the self-inspection object and to be checked for hidden dangers, the set of information to be checked for hidden dangers is information corresponding to an area to be checked for hidden dangers entered by the self-inspection object, the information to be checked for hidden dangers in the set of information to be checked for hidden dangers includes: a set of description information of the area to be checked for hidden dangers and a location of the area to be checked for hidden dangers, and the information types of the description information of the area to be checked for hidden dangers include: an image type, a text type, and a video type; a hidden danger identification unit, configured to identify hidden dangers for each hidden danger inspection information in the hidden danger inspection information set to generate hidden danger warning information, wherein the hidden danger warning information includes: hidden danger type identification, hidden danger risk level and hidden danger rectification suggestion information, and the hidden danger type identification indicates whether there is a hidden danger in the hidden danger inspection area corresponding to the hidden danger inspection information; A generating unit is configured to generate a hidden danger checking logic tree according to the hidden danger checking information set and the hidden danger warning information corresponding to the hidden danger checking information; The hidden danger handling unit is configured to handle hidden dangers at the target location according to the hidden danger investigation logic tree, wherein: Performing hidden danger identification on each hidden danger inspection information in the hidden danger inspection information set to generate hidden danger warning information, including: Activate the multimodal feature extraction block in the hidden danger warning information generation model that corresponds to the information type of the description information of the area to be checked for hidden dangers included in the information to be checked for hidden dangers, and obtain the activated hidden danger warning information generation model, wherein the hidden danger warning information generation model includes: 3 multimodal feature extraction blocks, 7 feature fusion blocks, 1 hidden danger type classifier, 1 hidden danger risk level classifier and hidden danger rectification suggestion information prediction head, the multimodal feature extraction blocks correspond to the information type of the description information of the area to be checked for hidden dangers one by one, and the hidden danger type classifier is a binary classifier; According to the activated hidden danger warning information generation model and the hidden danger inspection area description information set included in the hidden danger inspection information, hidden danger warning information corresponding to the hidden danger inspection information is generated, wherein: The generating of a hidden danger checking logic tree according to the hidden danger checking information set and the hidden danger warning information corresponding to the hidden danger checking information includes: Clustering the set of information to be checked for hidden dangers according to the position of the area to be checked for hidden dangers included in the information to be checked for hidden dangers in the set of information to be checked for hidden dangers to obtain a sequence of information groups to be checked for hidden dangers; An initial hidden danger troubleshooting logic tree is constructed with the target hidden danger troubleshooting information group, wherein the initial hidden danger troubleshooting logic tree includes: a root node and a child node set, the target hidden danger troubleshooting information group is the hidden danger troubleshooting information group located at the median position in the hidden danger troubleshooting information group sequence, and the number of child nodes in the child node set + 1 is the same as the number of target hidden danger troubleshooting information in the target hidden danger troubleshooting information group; According to the information groups to be checked for hidden dangers in the sequence of information groups to be checked for hidden dangers, excluding the target information group to be checked for hidden dangers, the initial hidden danger checking logic tree is updated to generate an initial hidden danger checking logic tree after tree update; According to the hidden danger risk level included in the hidden danger warning information corresponding to the hidden danger inspection information, the tree structure of the initial hidden danger inspection logic tree after the tree update is adjusted to obtain the hidden danger inspection logic tree.
7. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.
8. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.
