An information-based storage management device and method for tourist attractions

Abstract

The present invention discloses an information-based storage management device and method for tourist scenic spots, which relates to the technical field of storage management. After collecting left and right images through a binocular camera, face region detection is performed to obtain a left face image and a right face image. A target disparity map is obtained based on the left face image and the right face image, and if the depth change determined according to the target disparity map is greater than a preset change, feature extraction is performed on the left face image and the right face image to obtain target features. The corresponding storage cabinet is determined to be opened according to the target features. By performing face image recognition on the images collected by the binocular camera to obtain the face region, and then determining the disparity map through the face images to finally obtain the depth change, it can effectively judge whether the face recognition object is a real person or a photo, improving security. Finally, by extracting and matching the face depth features, the corresponding storage cabinet number is accurately found and the unlocking operation is executed, simplifying the access process and ensuring security.

Description

Technical Field

[0001] The present invention belongs to the technical field of storage management, and particularly relates to an information-based storage management device and method for tourist attractions. Background Art

[0002] Existing storage cabinets in tourist attractions have many deficiencies in actual applications and are difficult to meet the current needs of tourists and scenic area management. The identity verification methods of traditional storage cabinets usually rely on simple passwords, vouchers, or card swiping. These methods are not secure and reliable enough in the crowded environment of scenic areas. Tourists may face difficulties in retrieving their items due to losing vouchers or forgetting passwords. Moreover, the password method is easy to be peeked or tampered with by others. Once the voucher is lost or the password is leaked, there is a possibility that the stored items will be misretrieved or stolen, and the personal property safety of tourists cannot be fully guaranteed.

[0003] In order to improve convenience and security, many scenic areas have tried to introduce face recognition technology to replace the traditional identity verification method and perform access verification through faces. However, face recognition also faces new problems in actual applications, especially the risk of photo attacks. Simple face recognition systems are easily deceived by static photos, videos, or forged images, resulting in false unlocking or illegal item retrieval. Although face recognition simplifies the access process to a certain extent, it also brings security risks. Summary of the Invention

[0004] The purpose of the present invention is to solve the above problems and propose an information-based storage management device and method for tourist attractions.

[0005] In the first aspect of the implementation of the present invention, an information-based storage management method for tourist attractions is first proposed. The method includes:

[0006] When a retrieval instruction is received, a binocular camera is activated to collect images in the target area to obtain a left image and a right image;

[0007] The face regions of the left image and the right image are respectively detected to obtain a first target region and a second target region. The left image is cut according to the first target region to obtain a left face image, and the right image is cut according to the second target region to obtain a right face image;

[0008] A target disparity map is obtained based on the left face image and the right face image, the target disparity map is subjected to depth change processing to obtain a target depth image, and the depth change of the target depth image is extracted;

[0009] If the depth change is greater than a preset change, feature extraction is performed on the left face image and the right face image to obtain target features;

[0010] Find the storage cabinet number by searching the preset database according to the target feature, and open the corresponding storage cabinet according to the storage cabinet number.

[0011] Optionally, detecting the face regions of the left image and the right image respectively to obtain a first target region and a second target region includes:

[0012] For the initial face image, preprocess the initial face image to obtain a preprocessed image, and perform image enhancement on the preprocessed image to obtain a target image;

[0013] Substitute the target image into the preset face detection model to obtain a target detection frame; if the initial face image is the left image, the target detection frame is the first target region; if the initial face image is the right image, the target detection frame is the second target region.

[0014] Optionally, obtaining the target disparity map according to the left face image and the right face image includes:

[0015] Perform key point recognition on the left face image and the right face image respectively through the key point detection algorithm to obtain a first key point set and a second key point set;

[0016] Match the key points in the first key point set and the second key point set to obtain a key point combination set, and for each key point combination, obtain the disparity value of the key point combination through the semi-global block matching algorithm;

[0017] Substitute the left face image and the right face image into the real-time stereo matching network respectively to obtain a global disparity map, and correct the global disparity map according to all the key point combination disparity values to obtain a target disparity map.

[0018] Optionally, extracting features from the left face image and the right face image to obtain target features includes:

[0019] For the left face image and the right face image, segment the left face image and the right face image through the target grid to obtain a left face block set and a right face block set;

[0020] For each block in the left face block set and the right face block set, extract features from the block to obtain block features;

[0021] Generate a left face histogram according to all the block features in the left face block set, and generate a right face histogram according to all the block features in the right face block set;

[0022] Concatenate the left face histogram and the right face histogram to obtain a target histogram, and substitute the left face histogram, the right face histogram, and the target histogram into a preset feature fusion model to obtain target features.

[0023] Optionally, the method further includes:

[0024] When receiving a storage instruction, turn on the binocular camera to collect images in the target area to obtain a first left image and a first right image, and obtain item information;

[0025] Perform face area detection on the first left image and the first right image respectively to obtain a first target acquisition area and a second target acquisition area, cut the first left image according to the first target acquisition area to obtain a left face acquisition image, and cut the first right image according to the second target acquisition area to obtain a right face acquisition image;

[0026] Substitute the left face acquisition image and the right face acquisition image into the preset model to obtain target features, and find the express cabinet according to the item information to obtain the express cabinet number;

[0027] Bind the express cabinet number and the target features and save them to the preset database.

[0028] In the second aspect of the implementation of the present invention, an information-based storage management device for a tourist scenic area is proposed, including:

[0029] A pick-up image acquisition module, configured to turn on the binocular camera to collect images in the target area to obtain a left image and a right image when receiving a pick-up instruction;

[0030] A face image acquisition module, configured to perform face area detection on the left image and the right image respectively to obtain a first target area and a second target area, cut the left image according to the first target area to obtain a left face image, and cut the right image according to the second target area to obtain a right face image;

[0031] A depth change determination module, configured to obtain a target disparity map according to the left face image and the right face image, perform depth change processing on the target disparity map to obtain a target depth image, and extract the depth change of the target depth image;

[0032] A target feature determination module, configured to perform feature extraction on the left face image and the right face image to obtain target features if the depth change is greater than a preset change;

[0033] A storage cabinet number search module, configured to find the storage cabinet number from the preset database according to the target features, and open the corresponding storage cabinet according to the storage cabinet number.

[0034] Optionally, the face image acquisition module includes:

[0035] An image enhancement module, configured to perform preprocessing on the initial face image to obtain a preprocessed image, and perform image enhancement on the preprocessed image to obtain a target image;

[0036] A target detection box determination module, configured to substitute the target image into a preset face detection model to obtain a target detection box; if the initial face image is the left image, the target detection box is the first target area; if the initial face image is the right image, the target detection box is the second target area.

[0037] Optionally, the depth change determination module includes:

[0038] A key point recognition module, configured to perform key point recognition on the left face image and the right face image respectively through a key point detection algorithm to obtain a first key point set and a second key point set;

[0039] A key point matching module, configured to match the key points in the first key point set and the second key point set to obtain a key point combination set, and for each key point combination, obtain the disparity value of the key point combination through a semi-global block matching algorithm;

[0040] A target disparity map determination module, configured to substitute the left face image and the right face image into a real-time stereo matching network respectively to obtain a global disparity map, and correct the global disparity map according to all the key point combination disparity values to obtain a target disparity map.

[0041] Optionally, the target feature determination module includes:

[0042] An image segmentation module, configured to segment the left face image and the right face image through a target grid to obtain a left face square set and a right face square set for the left face image and the right face image;

[0043] A square feature extraction module, configured to extract features of each square in the left face square set and the right face square set to obtain square features;

[0044] A histogram generation module, configured to generate a left face histogram according to all the square features in the left face square set, and generate a right face histogram according to all the square features in the right face square set;

[0045] A histogram feature fusion module, which is used to splice the left face histogram and the right face histogram to obtain a target histogram, and substitute the left face histogram, the right face histogram, and the target histogram into a preset feature fusion model to obtain target features.

[0046] Optionally, the method further includes:

[0047] A storage image acquisition module, which is used to turn on a binocular camera to acquire images within a target area to obtain a first left image and a first right image when a storage instruction is received, and obtain item information;

[0048] A second face image acquisition module, which is used to perform face area detection on the first left image and the first right image respectively to obtain a first target acquisition area and a second target acquisition area, cut the first left image according to the first target acquisition area to obtain a left face acquisition image, and cut the first right image according to the second target acquisition area to obtain a right face acquisition image;

[0049] A locker number generation module, which is used to substitute the left face acquisition image and the right face acquisition image into the preset model to obtain target features, and find a locker according to the item information to obtain a locker number;

[0050] A data storage module, which is used to bind the locker number and the target features and store them in the preset database.

[0051] Advantages of the present invention:

[0052] The present invention provides an information-based storage management method for a tourist scenic area. When a fetching instruction is received, a binocular camera is turned on to acquire images within a target area to obtain a left image and a right image; face area detection is performed on the left image and the right image respectively to obtain a first target area and a second target area, the left image is cut according to the first target area to obtain a left face image, and the right image is cut according to the second target area to obtain a right face image; a target disparity map is obtained according to the left face image and the right face image, depth change processing is performed on the target disparity map to obtain a target depth image, and the depth change of the target depth image is extracted; if the depth change is greater than a preset change, feature extraction is performed on the left face image and the right face image to obtain target features; a storage cabinet number is found according to the target features in a preset database, and the corresponding storage cabinet is opened according to the storage cabinet number. By performing face image recognition on the images acquired by the binocular camera to obtain the face area, the data processing amount is reduced and the detection speed is increased. Then, the disparity map is determined through the face images, and finally the depth change is obtained, which can effectively determine whether the face recognition object is a real person or a photo, improving security. Finally, by extracting and matching the face depth features, the corresponding storage cabinet number is accurately found and the unlocking operation is performed, simplifying the storage and retrieval process and ensuring security. Brief Description of the Drawings

[0053] The present invention will be further described below with reference to the accompanying drawings.

[0054] Figure 1 FIG. is a flowchart of an information-based storage management method for a tourist scenic area provided by an embodiment of the present invention;

[0055] Figure 2 FIG. is a schematic structural diagram of an information-based storage management device for a tourist scenic area provided by an embodiment of the present invention. Detailed Embodiment

[0056] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The term "and / or" in this article is only a description of the associated relationship of the associated objects, indicating that there can be three relationships. For example, A and B can represent: A exists alone, A and B exist simultaneously, and B exists alone. These three situations. In addition, the descriptions such as "first" and "second" in the present invention are only for the purpose of description, and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the fact that those skilled in the art can implement them. When the combination of the technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of the technical solutions does not exist and is not within the protection scope required by the present invention.

[0057] All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0058] An embodiment of the present invention provides an information-based storage management method for a tourist scenic area. Refer to Figure 1 , Figure 1 FIG. is a flowchart of an information-based storage management method for a tourist scenic area provided by an embodiment of the present invention. The method includes the following steps:

[0059] S101, when a pick-up instruction is received, turn on the binocular camera to collect images in the target area to obtain a left image and a right image;

[0060] S102, respectively perform face area detection on the left image and the right image to obtain a first target area and a second target area, cut the left image according to the first target area to obtain a left face image, and cut the right image according to the second target area to obtain a right face image;

[0061] S103. Obtain a target disparity map based on the left face image and the right face image, perform depth change processing on the target disparity map to obtain a target depth image, and extract the depth change of the target depth image;

[0062] S104. If the depth change is greater than a preset change, perform feature extraction on the left face image and the right face image to obtain target features;

[0063] S105. Search a preset database according to the target features to obtain a storage cabinet number, and open the corresponding storage cabinet according to the storage cabinet number.

[0064] Based on an information-based storage management method for tourist scenic areas provided by an embodiment of the present invention, by performing face image recognition on the images collected by a binocular camera to obtain a face area, the data processing volume is reduced and the detection speed is increased. Then, a disparity map is determined through the face image, and finally the depth change is obtained, which can effectively determine whether the face recognition object is a real person or a photo, improving security. Finally, by extracting and matching the face depth features, the corresponding storage cabinet number is accurately found and the unlocking operation is executed, simplifying the access process and ensuring security.

[0065] In one implementation, by using a binocular camera to obtain images of a target area, a left-right disparity map can be constructed, thereby realizing the extraction of depth information, which helps to accurately locate the face position and reduce the false trigger rate; face feature extraction is only started after it is detected that the depth change reaches a specific threshold, which helps to save computing resources and perform recognition and matching only when the conditions are met, improving the response efficiency of the system.

[0066] In one implementation, the combination of face recognition and depth information can improve the accuracy of recognition and reduce the possibility of misrecognition. Feature extraction and database matching further strengthen the reliability of user recognition, thereby reducing the risk of unauthorized access; the target area is the area where a binocular camera can collect faces, and the target area will prompt tourists to stand at a specified position.

[0067] In one implementation, if the depth change is less than or equal to the preset change, it indicates that the shape of the detection object is relatively flat and there is no significant three-dimensional feature change, and it may not be a real person but a photo. In this case, subsequent detection processing is stopped.

[0068] In one implementation, the binocular camera only collects images when an item retrieval instruction is obtained, reducing resource consumption, and users do not need to perform related operations such as entering passwords, simplifying the access process; the preset change is determined by technical personnel, and the data stored in the preset database is the data generated when storing items.

[0069] In one embodiment, performing face area detection on the left image and the right image respectively to obtain a first target area and a second target area includes:

[0070] For the initial face image, preprocess the initial face image to obtain a preprocessed image, and perform image enhancement on the preprocessed image to obtain a target image;

[0071] Substitute the target image into a preset face detection model to obtain a target detection box; if the initial face image is a left image, the target detection box is the first target area; if the initial face image is a right image, the target detection box is the second target area.

[0072] In one implementation, image preprocessing and image enhancement can improve the image quality, making blurred, poorly lit, or low-contrast images clear, thereby improving the recognition accuracy of subsequent face detection models, especially having a significant effect under complex lighting or low-resolution conditions; the delineation of the target detection box enables the accurate recognition and marking of the face regions in the left and right images, reducing false detections or missed detections, which helps with disparity analysis in subsequent stereo vision calculations and ensures the reliability of depth information calculation.

[0073] In one implementation, after preprocessing and enhancing the image quality, the detection model can more effectively identify features, thereby improving the detection speed. Optimizing the process reduces the need for complex models or multiple detections, saving resources.

[0074] In one implementation, preprocessing is basic format conversion, normalization, etc.; image enhancement is to perform low-light enhancement on the left image and the right image respectively using Enhance-Net, enhancing the lighting information and details in the image, generating the enhanced left image and right image, thereby reducing the loss of details caused by low light and providing a clearer visual basis for subsequent face region detection.

[0075] In one implementation, the preset face detection model can be a face detection model such as RetinaFace, MTCNN, etc.

[0076] In one embodiment, obtaining the target disparity map from the left face image and the right face image includes:

[0077] Use the key point detection algorithm to perform key point recognition on the left face image and the right face image respectively to obtain the first key point set and the second key point set;

[0078] Match the key points in the first key point set and the second key point set to obtain a key point combination set. For each key point combination, obtain the disparity value of the key point combination through the semi-global block matching algorithm;

[0079] Substitute the left face image and the right face image into the real-time stereo matching network respectively to obtain the global disparity map, and correct the global disparity map according to all the key point combination disparity values to obtain the target disparity map.

[0080] In one implementation, the key point matching algorithm can accurately find the corresponding feature points in the left and right images, and accurately calibrate the depth information by calculating the disparity values of the key point combinations, effectively reducing the errors caused by factors such as illumination and noise; when calculating the key point disparity, the semi-global block matching algorithm can balance local and global information, reduce the matching errors caused by local occlusion and blur, etc., and increase the accuracy of stereo matching.

[0081] In one implementation, by using the key point matching results for the correction of the global disparity map, the accuracy of the disparity map can be greatly improved without sacrificing the overall efficiency, and at the same time, the dependence on complex stereo matching networks is reduced; by correcting the global disparity map with the disparity values of the key point combinations, the disparity results generated by the real-time stereo matching network can be calibrated, the mismatched areas caused by network errors can be reduced, and finally a more consistent target disparity map can be obtained.

[0082] In one implementation, the key point detection algorithm can be the Adaboost algorithm based on Haar features, OpenFace, dlib, SURF, etc.; specifically, for matching the key points in the first key point set and the second key point set to obtain the key point combination set, for each key point, a descriptor is generated using a feature descriptor algorithm such as SIFT, ORB, etc., which is used to characterize the local information of the point, and the most similar key point pair is found between the key point set of the left image and the key point set of the right image through the KNN matching algorithm, the Euclidean distance between the two descriptors is calculated, and the key point pair obtained by matching is taken as the one with the smallest Euclidean distance obtained.

[0083] In one implementation, the semi-global block matching algorithm can be SGM, Multi-level SGM, Adaptive SGM, Fast SGM, etc.; the real-time stereo matching network can be StereoNet, PSMNet, DeepPruner, Stereo-Unet, etc.

[0084] In one implementation, the global disparity map is corrected according to the disparity values of all key point combinations to obtain the target disparity map. Specifically, the difference in the disparity values of the key point combinations is obtained as the first difference, the difference corresponding to the key point combination in the global disparity map is obtained as the second difference, the difference between the first difference and the second difference is calculated as the correction difference, the pixel distance between the key point combinations is obtained, and the disparity values within each step length are corrected with the preset step length as the gradient. For the correction difference of each key point combination, within the pixel distance between them, the disparity values are gradually updated according to the preset step length. For example, at this time, the pixel distance between the key point combinations is 100 pixels, the key points are denoted as point A and point B, the preset step length is 20 pixel points, and the correction difference is 1. Then, starting from point A to point B, for the first 1 to 20 pixels, 0.2 is added to the formed disparity value, for the first 21 to 40 pixels, 0.4 is added to the formed disparity value, for the first 41 to 60 pixels, 0.6 is added to the formed disparity value, for the first 61 to 80 pixels, 0.8 is added to the formed disparity value, and for the first 81 to 100 pixels, 1 is added to the formed disparity value.

[0085] In one embodiment, the feature extraction of the left face image and the right face image to obtain the target features includes:

[0086] For the left face image and the right face image, the left face image and the right face image are segmented through the target grid to obtain the left face square set and the right face square set;

[0087] For each square in the left face square set and the right face square set, feature extraction is performed on the square to obtain the square features;

[0088] The left face histogram is generated according to all the square features in the left face square set, and the right face histogram is generated according to all the square features in the right face square set;

[0089] The left face histogram and the right face histogram are spliced to obtain the target histogram, and the left face histogram, the right face histogram, and the target histogram are substituted into the preset feature fusion model to obtain the target features.

[0090] In one implementation, by performing grid segmentation and block feature extraction on a face image, each block represents local information of the image. Such local features can capture image details such as texture and shape, thereby being able to more accurately represent different parts of the face. By generating histograms of the left and right faces and then performing splicing and feature fusion, these local features can be effectively combined to obtain a more comprehensive and discriminative target feature. By dividing the left and right face images into multiple blocks and extracting features for each block, the matching accuracy of facial features can be improved. Traditional holistic feature extraction methods may ignore changes in local regions of the image, while the block-based method can targetedly optimize features, thereby reducing the situation of false matches.

[0091] In one implementation, splicing the histograms of the left and right faces can better fuse visual information from the left and right perspectives. Since the left and right images usually contain different perspective information, feature fusion can integrate the advantages of both to improve the comprehensiveness and expressive ability of the features. The target feature can not only more accurately describe the facial features of an individual but also fuse detailed information from different perspectives.

[0092] In one implementation, through target grid segmentation and local feature extraction, the system can more flexibly adapt to complex environments. In some dynamic scenarios, different regions of the image may have different backgrounds, lighting, or interference factors. By extracting features of local blocks, these interferences can be reduced, the recognition ability of objects or faces can be improved, and thus the stability and speed of recognition can be enhanced.

[0093] In one implementation, the size of the target grid is determined by technicians and is usually a 36-by-36 pixel grid. The specific method for extracting block features from this block is to apply the LBP operator to each image block to extract the local texture features of each small block. The method calculates a binary number by comparing the gray value of each pixel with the gray values of its neighboring pixels, and then converts the binary number to a decimal number to obtain the feature value corresponding to this pixel point. The block feature is obtained by acquiring the feature values corresponding to all pixel points.

[0094] In one implementation, since the left and right images are obtained from two different angles, they contain different perspective information. During feature extraction, local texture features should be extracted from the left image and the right image respectively, and then these features are combined to form a complete feature vector. Specifically, the block features of the left image and the right image are spliced. For example, the block features of the first row in the left image and the block features of the first row in the right image are arranged in sequence to form a longer block feature, and the splicing is performed for each row in turn to obtain the target histogram. This will combine the feature information from the two perspectives and help improve the recognition accuracy, especially under different perspectives and poses.

[0095] In one implementation, substituting the left face histogram, the right face histogram, and the target histogram into a preset feature fusion model to obtain the target feature specifically involves performing scale feature extraction on the left face histogram, the right face histogram, and the target histogram respectively at the target scale to obtain a first left face scale map, a first right face scale map, and a first target face scale map, where the scales of the first left face scale map, the first right face scale map, and the first target face scale map are the same. Superposing the first left face scale map and the first right face scale map on the first target face scale map respectively to obtain a first left face fusion map and a first right face fusion map. Performing scale feature extraction on the first left face fusion map, the first right face fusion map, and the target histogram respectively at a second target scale to obtain a second left face scale map, a second right face scale map, and a second target face scale map. Then, obtaining a third left face scale map and a third right face scale map according to the same steps above. Repeating this step until feature extraction is performed at the final target scale to obtain a final left face scale map and a final right face scale map. Superposing the final left face scale map and the final right face scale map to obtain a final map, and performing average pooling on the final map and then passing it through a sigmoid function to obtain the target feature; the target scale and the final target scale are determined by technicians, and from the first target scale to the final target scale, the scale sizes decrease sequentially.

[0096] In one embodiment, the method further includes:

[0097] When a storage instruction is received, turn on the binocular camera to collect images in the target area to obtain a first left image and a first right image, and obtain item information;

[0098] Perform face region detection on the first left image and the first right image respectively to obtain a first target acquisition area and a second target acquisition area. Cut the first left image according to the first target acquisition area to obtain a left face acquisition image, and cut the first right image according to the second target acquisition area to obtain a right face acquisition image;

[0099] Substitute the left face acquisition image and the right face acquisition image into a preset model to obtain the target feature, and find the express cabinet number according to the item information;

[0100] Bind the express cabinet number and the target feature and save them to a preset database.

[0101] In one implementation, by binding the features of the item with the express cabinet number and combining with the storage in the preset database, tracking and recording of each item can be formed in the system. Managers can easily query the specific location where the item is stored, which helps to improve the efficiency of item management. The item information is the information input by the user, such as the size and weight of the item.

[0102] Embodiments of the present invention also provide an information-based storage management device for tourist attractions based on the same inventive concept. Refer to Figure 2 , Figure 2 which is a schematic structural diagram of an information-based storage management device for tourist attractions provided by an embodiment of the present invention, including:

[0103] An item retrieval image acquisition module, configured to turn on a binocular camera to acquire images in a target area when receiving an item retrieval instruction, obtaining a left image and a right image;

[0104] A face image acquisition module, configured to respectively perform face area detection on the left image and the right image to obtain a first target area and a second target area, cut the left image according to the first target area to obtain a left face image, and cut the right image according to the second target area to obtain a right face image;

[0105] A depth change determination module, configured to obtain a target disparity map based on the left face image and the right face image, perform depth change processing on the target disparity map to obtain a target depth image, and extract the depth change of the target depth image;

[0106] A target feature determination module, configured to perform feature extraction on the left face image and the right face image to obtain target features if the depth change is greater than a preset change;

[0107] A storage cabinet number search module, configured to search a preset database according to the target features to obtain a storage cabinet number, and open the corresponding storage cabinet according to the storage cabinet number.

[0108] Based on the information-based storage management device for tourist attractions provided by an embodiment of the present invention, by performing face image recognition on the images acquired by the binocular camera to obtain the face area, the data processing volume is reduced and the detection speed is increased. Then, the disparity map is determined through the face image to finally obtain the depth change, which can effectively determine whether the face recognition object is a real person or a photo, improving the security. Finally, by extracting and matching the face depth features, the corresponding storage cabinet number is accurately found and the unlocking operation is executed, simplifying the access process and ensuring the security.

[0109] In one embodiment, the face image acquisition module includes:

[0110] An image enhancement module, configured to perform preprocessing on an initial face image to obtain a preprocessed image, and perform image enhancement on the preprocessed image to obtain a target image for the initial face image;

[0111] A target detection box determination module, configured to substitute the target image into a preset face detection model to obtain a target detection box; if the initial face image is the left image, the target detection box is the first target area; if the initial face image is the right image, the target detection box is the second target area.

[0112] In one embodiment, the depth change determination module includes:

[0113] A key point recognition module, configured to respectively perform key point recognition on the left face image and the right face image through a key point detection algorithm to obtain a first key point set and a second key point set;

[0114] A key point matching module, configured to match the key points in the first key point set and the second key point set to obtain a key point combination set, and for each key point combination, obtain the disparity value of the key point combination through a semi-global block matching algorithm;

[0115] A target disparity map determination module, configured to respectively substitute the left face image and the right face image into a real-time stereo matching network to obtain a global disparity map, and correct the global disparity map according to all the key point combination disparity values to obtain a target disparity map.

[0116] In one embodiment, the target feature determination module includes:

[0117] An image segmentation module, configured to segment the left face image and the right face image through a target grid for the left face image and the right face image to obtain a left face square set and a right face square set;

[0118] A square feature extraction module, configured to extract features from each square in the left face square set and the right face square set to obtain square features;

[0119] A histogram generation module, configured to generate a left face histogram according to all the square features in the left face square set, and generate a right face histogram according to all the square features in the right face square set;

[0120] A histogram feature fusion module, configured to splice the left face histogram and the right face histogram to obtain a target histogram, and substitute the left face histogram, the right face histogram, and the target histogram into a preset feature fusion model to obtain a target feature.

[0121] In one embodiment, the method further includes:

[0122] A storage image acquisition module, configured to turn on a binocular camera to acquire images in a target area to obtain a first left image and a first right image and obtain item information when receiving a storage instruction;

[0123] A second face image acquisition module, configured to respectively perform face area detection on the first left image and the first right image to obtain a first target acquisition area and a second target acquisition area, cut the first left image according to the first target acquisition area to obtain a left face acquisition image, and cut the first right image according to the second target acquisition area to obtain a right face acquisition image;

[0124] The express cabinet number generation module is used to substitute the left face capture image and the right face capture image into a preset model to obtain target features, and find the express cabinet according to the item information to obtain the express cabinet number;

[0125] The data storage module is used to bind the express cabinet number and the target features and store them in a preset database.

[0126] The above has described an embodiment of the present invention in detail, but the content is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the application of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. An information-based storage management method for tourist attractions, characterized in that The method includes: When receiving an item-taking instruction, turn on the binocular camera to collect images in the target area to obtain a left image and a right image; Detect the face regions in the left image and the right image respectively to obtain a first target region and a second target region. Cut the left image according to the first target region to obtain a left face image, and cut the right image according to the second target region to obtain a right face image; Obtain a target disparity map based on the left face image and the right face image, perform depth change processing on the target disparity map to obtain a target depth image, and extract the depth change of the target depth image; If the depth change is greater than a preset change, extract features from the left face image and the right face image to obtain target features; Search for a storage cabinet number in a preset database according to the target features, and open the corresponding storage cabinet according to the storage cabinet number; Extracting features from the left face image and the right face image to obtain target features includes: For the left face image and the right face image, use a target grid to segment the left face image and the right face image to obtain a left face block set and a right face block set; For each block in the left face block set and the right face block set, extract features from the block to obtain block features; Generate a left face histogram based on all the block features in the left face block set, and generate a right face histogram based on all the block features in the right face block set; Stitch the left face histogram and the right face histogram to obtain a target histogram, and substitute the left face histogram, the right face histogram, and the target histogram into a preset feature fusion model to obtain target features; The size of the target grid is a 36 by 36 pixel grid. Extracting features from the block to obtain block features specifically means: applying the LBP operator to each image block, extracting the local texture features of each small block, calculating a binary number by comparing the gray value of each pixel with that of its neighboring pixels, and then converting the binary to decimal to obtain the feature value corresponding to the pixel point, and obtaining the block features by acquiring the feature values corresponding to all pixel points; Substituting the left face histogram, the right face histogram, and the target histogram into the preset feature fusion model to obtain the target feature specifically includes: performing scale feature extraction on the left face histogram, the right face histogram, and the target histogram respectively at the target scale to obtain the first left face scale map, the first right face scale map, and the first target face scale map, where the scales of the first left face scale map, the first right face scale map, and the first target face scale map are the same; superimposing the first left face scale map and the first right face scale map on the first target face scale map respectively to obtain the first left face fusion map and the first right face fusion map; performing scale feature extraction on the first left face fusion map, the first right face fusion map, and the target histogram respectively at the second target scale to obtain the second left face scale map, the second right face scale map, and the second target face scale map, and then obtaining the third left face scale map and the third right face scale map according to the same steps until the final left face scale map and the final right face scale map are obtained by performing feature extraction at the final target scale; superimposing the final left face scale map and the final right face scale map to obtain the final map, and performing average pooling on the final map and then passing it through the sigmoid function to obtain the target feature.

2. The information-based storage management method for a tourist scenic area according to claim 1, characterized in that Performing face region detection on the left image and the right image respectively to obtain the first target region and the second target region includes: For the initial face image, preprocessing the initial face image to obtain a preprocessed image, and performing image enhancement on the preprocessed image to obtain the target image; Substituting the target image into the preset face detection model to obtain the target detection box; if the initial face image is the left image, the target detection box is the first target region; if the initial face image is the right image, the target detection box is the second target region.

3. The information-based storage management method for a tourist scenic area according to claim 1, characterized in that, Obtaining the target disparity map according to the left face image and the right face image includes: Performing key point recognition on the left face image and the right face image respectively through the key point detection algorithm to obtain the first key point set and the second key point set; Matching the key points in the first key point set and the second key point set to obtain the key point combination set, and for each key point combination, obtaining the disparity value of the key point combination through the semi-global block matching algorithm; Substituting the left face image and the right face image into the real-time stereo matching network respectively to obtain the global disparity map, and correcting the global disparity map according to all the disparity values of the key point combinations to obtain the target disparity map.

4. The information-based storage management method for a tourist scenic area according to claim 1, characterized in that, The method further includes: When receiving the storage instruction, turning on the binocular camera to collect the images in the target region to obtain the first left image and the first right image, and obtaining the item information; Performing face region detection on the first left image and the first right image respectively to obtain the first target collection region and the second target collection region, cutting the first left image according to the first target collection region to obtain the left face collection image, and cutting the first right image according to the second target collection region to obtain the right face collection image. Substitute the left face acquisition image and the right face acquisition image into a preset feature fusion model to obtain target features, and find a locker according to the item information to obtain a locker number; After binding the locker number and the target features, save them to the preset database.

5. An information-based storage management device for a tourist scenic area, which is used to implement the information-based storage management method for a tourist scenic area according to any one of claims 1-4, and is characterized in that, The device includes: An item-taking image acquisition module, configured to turn on a binocular camera to acquire images in a target area when receiving an item-taking instruction, obtaining a left image and a right image; A face image acquisition module, configured to perform face area detection on the left image and the right image respectively to obtain a first target area and a second target area, cut the left image according to the first target area to obtain a left face image, and cut the right image according to the second target area to obtain a right face image; A depth change determination module, configured to obtain a target disparity map according to the left face image and the right face image, perform depth change processing on the target disparity map to obtain a target depth image, and extract the depth change of the target depth image; A target feature determination module, configured to, if the depth change is greater than a preset change, perform feature extraction on the left face image and the right face image to obtain target features; A locker number search module, configured to find a locker number from a preset database according to the target features, and open the corresponding storage cabinet according to the locker number.

6. The information-based storage management device for a tourist scenic area according to claim 5, characterized in that, The face image acquisition module includes: An image enhancement module, configured to perform preprocessing on an initial face image to obtain a preprocessed image, and perform image enhancement on the preprocessed image to obtain a target image; A target detection box determination module, configured to substitute the target image into a preset face detection model to obtain a target detection box; if the initial face image is the left image, the target detection box is the first target area; if the initial face image is the right image, the target detection box is the second target area.

7. An information-based storage management device for a tourist scenic area according to claim 5, characterized in that, The depth change determination module includes: A key point recognition module, configured to respectively perform key point recognition on the left face image and the right face image through a key point detection algorithm to obtain a first key point set and a second key point set; A key point matching module, configured to match the key points in the first key point set and the second key point set to obtain a key point combination set, and for each key point combination, obtain the disparity value of the key point combination through a semi-global block matching algorithm; A target disparity map determination module, configured to substitute the left face image and the right face image into a real-time stereo matching network respectively to obtain a global disparity map, and correct the global disparity map according to all the key point combination disparity values to obtain a target disparity map.

8. An information-based storage management device for a tourist scenic area according to claim 5, characterized in that, The target feature determination module includes: An image segmentation module, configured to segment the left face image and the right face image through a target grid for the left face image and the right face image to obtain a left face block set and a right face block set; A block feature extraction module, configured to extract features of each block in the left face block set and the right face block set to obtain block features; A histogram generation module, configured to generate a left face histogram according to all block features in the left face block set, and generate a right face histogram according to all block features in the right face block set; A histogram feature fusion module, configured to splice the left face histogram and the right face histogram to obtain a target histogram, and substitute the left face histogram, the right face histogram and the target histogram into a preset feature fusion model to obtain target features.

9. An information-based storage management device for tourist attractions according to claim 5, characterized in that, The device further includes: A storage image acquisition module, configured to turn on a binocular camera to acquire an image in a target area to obtain a first left image and a first right image when a storage instruction is received, and obtain item information; A second face image acquisition module, configured to respectively perform face area detection on the first left image and the first right image to obtain a first target acquisition area and a second target acquisition area, cut the first left image according to the first target acquisition area to obtain a left face acquisition image, and cut the first right image according to the second target acquisition area to obtain a right face acquisition image; A locker number generation module, configured to substitute the left face acquisition image and the right face acquisition image into a preset feature fusion model to obtain target features, and find a locker according to the item information to obtain a locker number; A data storage module, configured to bind the locker number and the target features and store them in the preset database.

Images