A human weight recognition system based on local features

Abstract

The present invention discloses a human weight recognition system based on local features, which relates to the technical field of human weight recognition. The system aims to solve the problem of information recognition, recording, and management. The system specifically includes an information acquisition module, a cloud database, an image comparison and recognition module, an object recognition log, and a security management module. The information acquisition module includes biometric feature acquisition, sound acquisition, and behavioral feature acquisition, and the cloud database includes data analysis and data management. The image comparison and recognition module includes image acceptance, preprocessing, feature extraction, feature selection, and analysis, comparison, and discrimination. The object recognition log includes log storage and log management, and the security management module includes recognition warning and information security encryption. By adding the object recognition log and security management module to the overall human weight recognition system, the present invention can effectively store and archive daily information logs. Furthermore, the marked documents in the information database facilitate users to retrieve problematic information features.

Description

Technical Field

[0001] The present invention relates to the technical field of human weight recognition, and in particular to a human weight recognition system based on local features. Background Art

[0002] Human weight recognition is used to identify known targets from video or still images. The current process generally involves first evaluating the image quality. Once the image quality meets the requirements, the features of the target object in the image are determined. These features are then compared with the features of known targets to determine whether the target object in the image is a known target.

[0003] In crowded places such as airports and subway stations, human feature-based re-identification technology has very important application value and has attracted the attention of more and more researchers.

[0004] After searching, the Chinese patent application number CN202110044449.8 discloses a method and device for human weight recognition. The method includes: establishing a face negative film library and a body negative film library based on a first video frame obtained by a first image acquisition device, and recording the correspondence between the person's identification and the face identification and / or the body identification; based on the second video frame obtained by a second image device, the face negative film library, the body negative film library, and the correspondence between the person's identification and the face identification and / or the body identification. A human weight recognition method and device in the above patent has the following deficiencies:

[0005] Although the overall device can identify and analyze human weight, it does not perform security management and daily log recording of the identified object information. As a result, the security of object information identification cannot be guaranteed, and it is relatively difficult to find related subsequent information. Summary of the Invention

[0006] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a human weight recognition system based on local features.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] A human weight recognition system based on local features includes an information collection module, a cloud database, an image comparison and recognition module, an object recognition log, and a security management module. The information collection module includes biometric feature collection, sound collection, and behavioral feature collection, and the cloud database includes data analysis and data management. The image comparison and recognition module includes image acceptance, preprocessing, feature extraction, feature selection, and analysis, comparison, and judgment. The object recognition log includes log storage and log management. The security management module includes recognition warning and information security encryption.

[0009] Preferably, the information collection module includes the following steps:

[0010] A1: The person to be identified enters the collection door and stands inside the collection door;

[0011] A2: The acquisition door uses cameras to perform temperature spectrum recognition, facial recognition, iris recognition, body feature recognition, voice feature recognition, and fingerprint recognition on the person to be identified to collect relevant image and text data;

[0012] A3: Next, the collected data will be archived and the data collection information will be transmitted to the cloud database for analysis and comparison.

[0013] Furthermore: the image comparison and recognition module includes the following steps:

[0014] B1: Image acceptance: accept the image information of the biometric features of the person to be identified and save it for archiving;

[0015] B2: Preprocessing: Encode and compress image information to reduce the amount of image data, enhance and restore the image to make the feature contours and details clear, and use filtering methods to ensure the image details;

[0016] B3: Feature extraction: Extract multiple local features from the processed image, divide and store each local feature, and send it to the cloud database for feature comparison and analysis;

[0017] B4: Feature selection: The cloud database will feed back the comparison results obtained through analysis, and select and extract the local features of the target in the archive;

[0018] B5: Analysis, comparison and discrimination: Perform feature analysis on the selected target local features and compare them with the local features in the cloud database to obtain the discrimination results.

[0019] Based on the above solution: the object identification log includes the following steps:

[0020] C1: Log storage: The daily human weight recognition information is archived and integrated, uploaded to the cloud database, and each sub-information database is filled daily;

[0021] C2: Log management: Integrate the feature information in each sub-information database, remove the same feature information and supplement the relevant feature information, and store the sub-information database securely.

[0022] A more preferred solution among the above solutions is that the security management module includes the following steps:

[0023] D1: Identification and warning: The feedback information is marked in red, saved and archived, and the relevant feature information is compared and identified;

[0024] D2: Information security encryption: The sub-information database is securely encrypted and marked as such.

[0025] As a further solution of the present invention: the collection door includes two main frames and a top plate, and the outer walls of the opposite sides of the two main frames are respectively provided with guide grooves, and slide grooves are provided on both sides of the two guide grooves. Three warning lights are provided on the outer wall of the main frame close to the guide groove, and a screw is provided on the inner wall of the main frame close to the guide groove, and the circumferential outer wall of the screw is sleeved with a fixed plate, and guide plates are respectively provided on the outer walls on both sides of the fixed plate, and the opposite sides of the two guide plates form a sliding fit with the slide groove, and a laser profile sensor is provided on the outer wall of one side of the fixed plate, and the top of the two screws is provided with the same gear transmission mechanism, and the middle bottom end of the gear transmission mechanism is fixed with an intermittent rotation mechanism through a U-shaped bracket, cameras are provided around the bottom of the intermittent rotation mechanism, and the intermittent rotation mechanism and the gear transmission mechanism are linked to each other, a machine box is provided on the top middle outer wall of the top plate, and a limiting mirror is provided on the outer wall of one side of the guide plate.

[0026] At the same time, the gear transmission mechanism includes two first main gears, two first gears, a rotating rod, a motor, a second main gear and a second gear, and the motor is arranged inside the chassis, the second main gear is connected to the output end of the motor through a connecting shaft, the rotating rod is fixed to the inner walls on both sides of the top plate through bearings, the second gear is sleeved on the middle circumference of the rotating rod, and the second main gear and the second gear are engaged with each other, the two first main gears are respectively arranged on the top circumference of the two lead screws, the two first gears are respectively sleeved on the two ends of the rotating rod, the first main gear and the first gear are engaged with each other, and the U-shaped bracket is connected to the circumference of the rotating rod through bearings.

[0027] As a preferred embodiment of the present invention: the intermittent rotation mechanism includes a linkage gear, a first drive gear, a second drive gear, a cam, a turntable and a fixed block, and the fixed block is arranged on the middle outer wall of the bottom end of the U-shaped bracket, and the top outer wall of the fixed block is provided with a groove, the first drive gear and the second drive gear are respectively rotatably connected to the bottom of the groove through a guide shaft, the first drive gear and the second drive gear are meshed with each other, and the linkage gear is arranged at the top end of the guide shaft in the first drive gear, the linkage gear and the second gear are meshed with each other, the cam is arranged at the bottom end of the guide shaft of the second drive gear, and the turntable is rotatably connected to the U-shaped bracket close to the bottom outer wall of the fixed block through a connecting rod, the top outer wall of the turntable is provided with meshing grooves distributed at equal distances, and the cam and the meshing grooves are adapted to each other, and the bottom outer wall of the turntable is connected to a cross guide frame through a fixed rod.

[0028] At the same time, a winding drum is provided on the outer wall of the main frame near the top of the guide groove, and an elastic sealing plate is connected to the winding drum through a ruler spring, a handle is provided at one end of the elastic sealing plate, and clamping protrusions are provided on both sides of the end of the elastic sealing plate, and a clamping seat is provided on one side of the bottom of the main frame, and the clamping protrusion is adapted to the clamping seat.

[0029] As a more preferred solution of the present invention: a mounting casing is provided on one side outer wall of the main frame, and a start-stop mechanism is provided inside the mounting casing.

[0030] The beneficial effects of the present invention are:

[0031] 1. This human weight recognition system based on local features can effectively store and archive daily information logs by adding object recognition logs and security management modules to the overall human weight recognition system, ensuring the richness of the information database. At the same time, the marked documents in the information database facilitate users to extract problematic information features, improving the efficiency of information security recognition. At the same time, the daily log information database is encrypted to ensure the security of personal information, avoid information leakage, and strengthen the management security of the recognition system.

[0032] 2. This human weight recognition system based on local features, when working, the person to be identified passes through two main frames and stands between the two main frames. At this time, the gear transmission mechanism is started, and the gear transmission mechanism drives the two screws to rotate, and the rotating screws promote the fixed plate and its two guide plates to move up and down in the slide groove. In the meantime, the laser contour sensor on one side of the fixed plate is started, and the lifting laser contour sensor adopts the laser triangulation reflection principle. The laser beam is amplified to form a laser line projected onto the surface of the human body. The reflected light passes through the high-quality optical system and is projected onto the imaging matrix. The distance from the sensor to the measured surface and the position information along the laser line are obtained through calculation. The three-dimensional contour of the measured target can be directly obtained through data processing. At the same time, the gear transmission mechanism drives the intermittent rotation mechanism to work, so that the multiple cameras at its bottom end can shoot the surroundings of the human body. In this way, the feature image acquisition of the human body is completed, the effective acquisition and uploading of the feature image of the human body is promoted, the recording of the human body's physical signs and appearance images is ensured, and the comprehensiveness of recognition is improved.

[0033] 3. This human weight recognition system based on local features starts the motor, the motor drives the second main gear to rotate, and promotes the rotation of the rotating rod through the second gear. At the same time, the rotating rotating rod promotes the rotation of the second gears at both ends of it, thereby driving the two lead screws in the main frame to rotate, and the two rotating lead screws drive the two laser profile sensors to perform lifting and lowering work, ensuring the quality of collecting human body features. When the rotating second gear drives the linkage gear to rotate, the linkage gear rotates the first drive gear. At the same time, through the relationship of meshing with the second drive gear, the cam at the bottom of the second drive gear is driven to rotate, and the cam is adapted to the meshing groove, thereby driving the turntable to rotate intermittently, and the intermittently rotating turntable effectively drives the camera of the cross guide frame at its bottom to rotate intermittently, which not only ensures that the camera can clearly collect human feature images during the intermittent process, but also ensures multi-angle shooting of human weight recognition images, thereby promoting recognition accuracy.

[0034] 4. This is a human body weight recognition system based on local features. When the entire device is not in use, the push buckle is squeezed upward to cause the inclined top of the push buckle to move upward, and the displaced inclined top passes through the inclined hole in the card block, causing the card block to slide horizontally in the guide groove. At this time, the spring on one side of the card block is tightened, and one end of the card block is disengaged from the limit card plate on the circumference of the turntable. At the same time, the elastic sealing plate wrapped around the winding roller is pulled out of the winding drum and moves vertically downward under the action of the two counterweights at its end. When the end of the elastic sealing plate falls to the bottom of the main frame, the two magnetic seats absorb the two counterweights, thereby completing the sealing effect of the elastic sealing plate on the inside of the device, reducing the workload of the staff while promoting the protection of the laser profile sensor and its components.

[0035] 5. This is a human weight recognition system based on local features. When the device is enabled, the pedal is pressed by the foot, causing one end of the pedal to tilt. At this time, the tilted end of the pedal triggers the pressing valve, causing the hydraulic system to work, and the hydraulic oil is introduced into the liquid inlet pipe. The hydraulic oil enters the hydraulic rotating assembly through the liquid inlet pipe, so that the connecting shaft at one end drives the turntable to rotate. At this time, the rotating turntable drives the winding roller to rotate through the connecting shaft, thereby causing the elastic sealing plate outside the winding drum to rise upward to perform the winding work. When the winding work is completed, the counterweight block on one side of the end of the elastic sealing plate sends the copper block into the electrified cavity, and the copper block now connects the two connecting paddles in the electrified cavity, so that electricity is supplied to the connecting wire, thereby driving the motor in the chassis to start working. Therefore, while lifting the elastic sealing plate, the startup action of the entire device is simplified, and automation is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of a human weight recognition system based on local features proposed by the present invention;

[0037] Figure 2 This is a schematic diagram of the main structure of a biometric feature collection gate in a human weight recognition system based on local features proposed by the present invention;

[0038] Figure 3 This is a schematic diagram of the internal structure of a biometric feature collection gate in a human weight recognition system based on local features proposed by the present invention;

[0039] Figure 4 This is a structural diagram of a biometric feature collection gate A in a human weight recognition system based on local features proposed by the present invention;

[0040] Figure 5 This is a structural diagram of the biometric feature collection gate B in a human weight recognition system based on local features proposed by the present invention;

[0041] Figure 6 This is a structural diagram of a housing installed in the main frame of a biometric feature collection door in a human weight recognition system based on local features proposed by the present invention;

[0042] Figure 7 This is a schematic diagram of the local structure inside the top of the housing installed in the biometric feature collection door in the human weight recognition system based on local features proposed by the present invention;

[0043] Figure 8 This is a schematic cross-sectional view of one side of the winding drum in a biometric feature collection door in a human weight recognition system based on local features proposed by the present invention.

[0044] In the figure: 1-main frame, 2-warning light, 3-guide groove, 4-top plate, 5-chassis, 6-cross guide, 7-camera, 8-first main gear, 9-first gear, 10-screw, 11-fixed plate, 12-guide plate, 13-rotating rod, 14-U-shaped bracket, 15-motor, 16-limiting mirror, 17-laser profile sensor, 18-slide, 19-winding drum, 20-handle, 21-second main gear, 22-second gear, 23-linkage gear, 2 4-first drive gear, 25-second drive gear, 26-cam, 27-turntable, 28-fixed block, 29-mounting housing, 30-press buckle, 31-liquid outlet pipe, 32-liquid inlet pipe, 33-press valve, 34-pedal, 35-magnetic seat, 36-counterweight, 37-hydraulic rotating assembly, 38-turntable, 39-limiting clamping plate, 40-block, 41-oblique hole, 42-spring, 43-oblique top, 44-copper block, 45-connection paddle, 46-connection wires. DETAILED DESCRIPTION

[0045] The technical solution of this patent is further described in detail below in conjunction with specific implementation methods.

[0046] The following describes in detail embodiments of the present invention, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0047] In the description of this patent, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings. They are only for the convenience of describing this patent and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, they should not be understood as limitations on this patent.

[0048] In the description of this patent, it should be noted that, unless otherwise expressly specified or limited, the terms "installed," "connected," "connected," and "set" should be understood in a broad sense. For example, they can refer to fixed connection or set, detachable connection or set, or integral connection or set. Those skilled in the art will understand the specific meanings of the above terms in this patent based on the specific circumstances.

[0049] Example 1:

[0050] A human weight recognition system based on local features, such as Figure 1 As shown, it includes an information collection module, a cloud database, an image comparison and recognition module, an object recognition log and a security management module. The information collection module includes biometric feature collection, sound collection and behavioral feature collection, and the cloud database includes data analysis and data management. The image comparison and recognition module includes image acceptance, preprocessing, feature extraction, feature selection and analysis and comparison judgment. The object recognition log includes log storage and log management. The security management module includes recognition warning and information security encryption.

[0051] In order to facilitate the collection of human information; Figure 1 As shown, the information collection module includes the following steps:

[0052] A1: The person to be identified enters the collection door and stands inside the collection door;

[0053] A2: The acquisition door uses cameras to perform temperature spectrum recognition, facial recognition, iris recognition, body feature recognition, voice feature recognition, and fingerprint recognition on the person to be identified to collect relevant image and text data;

[0054] A3: Next, the collected data will be archived and the data collection information will be transmitted to the cloud database for analysis and comparison.

[0055] In order to facilitate comparative analysis of received image information; e.g. Figure 1 As shown, the image comparison and recognition module includes the following steps:

[0056] B1: Image acceptance: accept the image information of the biometric features of the person to be identified and save it for archiving;

[0057] B2: Preprocessing: Encode and compress image information to reduce the amount of image data, enhance and restore the image to make the feature contours and details clear, and use filtering methods to ensure the image details;

[0058] B3: Feature extraction: Extract multiple local features from the processed image, divide and store each local feature, and send it to the cloud database for feature comparison and analysis;

[0059] B4: Feature selection: The cloud database will feed back the comparison results obtained through analysis, and select and extract the local features of the target in the archive;

[0060] B5: Analysis, comparison and discrimination: Perform feature analysis on the selected target local features and compare them with the local features in the cloud database to obtain the discrimination results.

[0061] In order to carry out security management and daily log recording of identification object information; Figure 1 As shown, object recognition logging includes the following steps:

[0062] C1: Log storage: The daily human weight recognition information is archived and integrated, uploaded to the cloud database, and each sub-information database is filled daily;

[0063] C2: Log management: Integrate the feature information in each sub-information database, remove the same feature information and supplement the relevant feature information, and store the sub-information database securely.

[0064] The security management module includes the following steps:

[0065] D1: Identification and warning: The feedback information is marked in red, saved and archived, and the relevant feature information is compared and identified;

[0066] D2: Information security encryption: The sub-information database is securely encrypted and marked as such.

[0067] By adding object recognition logs and security management modules to the overall human weight recognition system, daily information logs can be effectively stored and archived, ensuring the fullness of the information database. At the same time, the marked documents in the information database make it convenient for users to extract problematic information features, thereby improving the efficiency of information security identification. At the same time, the daily log information database is encrypted to ensure the security of personal information, avoid information leakage, and strengthen the security of identification system management.

[0068] Example 2:

[0069] A human weight recognition system based on local features, such as Figure 2-5 As shown, in order to strengthen the collection of human biometric images; this embodiment adds a corresponding collection device on the basis of embodiment 1, and the collection door includes two main frames 1 and a top plate 4, and the outer walls of the two main frames 1 on the opposite side are respectively provided with guide grooves 3, and both sides of the two guide grooves 3 are provided with slide grooves 18, and the outer wall of the main frame 1 close to the guide groove 3 is fixed with three warning lights 2 by bolts, and the inner wall of the main frame 1 close to the guide groove 3 is fixed with a screw 10 by a thread, and the circumferential outer wall of the screw 10 is sleeved with a fixing plate 11, and the outer walls on both sides of the fixing plate 11 are respectively fixed with guide plates 12 by bolts, and the two guide plates 12 The opposite sides form a sliding fit with the slide groove 18, the outer wall of one side of the fixed plate 11 is fixed with a laser profile sensor 17 by bolts, the outer wall of one side of the guide plate 12 is fixed with a limit mirror 16 by bolts, the top of the two screws 10 is provided with the same gear transmission mechanism, and the middle bottom end of the gear transmission mechanism is fixed with an intermittent rotation mechanism through a U-shaped bracket 14, the bottom of the intermittent rotation mechanism is provided with cameras 7, and the intermittent rotation mechanism and the gear transmission mechanism are linked to each other, the top middle outer wall of the top plate 4 is fixed with the chassis 5 by bolts, and the model of the laser profile sensor 17 is Gocator2300.

[0070] During operation, the person to be identified passes through the two main frames 1 and stands between the two main frames 1. At this time, the gear transmission mechanism is started, and the gear transmission mechanism drives the two screws 10 to rotate, and the rotating screws 10 promote the fixed plate 11 and its two guide plates 12 to move up and down in the slide 18. In the meantime, the laser profile sensor 17 on one side of the fixed plate 11 is started. The lifting laser profile sensor 17 adopts the laser triangulation reflection principle. The laser beam is amplified to form a laser line and projected onto the surface of the human body. The reflected light passes through the high-quality optical system and is projected onto the imaging matrix. The distance from the sensor to the measured surface and the position information along the laser line are calculated. The three-dimensional contour of the measured target can be directly obtained through data processing. At the same time, the gear transmission mechanism drives the intermittent rotation mechanism to work, so that the multiple cameras 7 at its bottom can take pictures of the human body all around. In this way, the feature image acquisition of the human body is completed, the effective acquisition and uploading of the feature image of the human body is promoted, the recording of the human body's physical features is ensured, and the comprehensiveness of the recognition is improved. The limiting mirror 16 on one side of the guide plate 12 ensures the scanning quality of the laser contour sensor 17, avoiding the influence of ambient light on the scanning.

[0071] The gear transmission mechanism includes two first main gears 8, two first gears 9, a rotating rod 13, a motor 15, a second main gear 21 and a second gear 22. The motor 15 is fixed to the inside of the chassis 5 by bolts. The second main gear 21 is connected to the output end of the motor 15 through a connecting shaft. The rotating rod 13 is fixed to the inner walls of both sides of the top plate 4 by bearings. The second gear 22 is sleeved on the middle circumference of the rotating rod 13, and the second main gear 21 and the second gear 22 are meshed with each other. The two first main gears 8 are respectively fixed to the top circumference of the two lead screws 10 by threads. The two first gears 21 and the second gear 22 are meshed with each other. The wheels 9 are respectively sleeved on the two ends of the rotating rod 13, the first main gear 8 and the first gear 9 are engaged with each other, and the U-shaped bracket 14 is connected to the circumference of the rotating rod 13 through a bearing; when working, the motor 15 is started, and the motor 15 drives the second main gear 21 to rotate, and promotes the rotation of the rotating rod 13 through the second gear 22. At the same time, the rotating rotating rod 13 promotes the rotation of the second gear 9 at both ends thereof, thereby driving the two lead screws 10 in the main frame 1 to rotate, and the two rotating lead screws 10 drive the two laser profile sensors 17 to perform lifting and lowering work, thereby ensuring the quality of collecting human body vital signs and appearance.

[0072] The intermittent rotation mechanism includes a linkage gear 23, a first drive gear 24, a second drive gear 25, a cam 26, a turntable 27 and a fixed block 28, and the fixed block 28 is fixed to the middle outer wall of the bottom end of the U-shaped bracket 14 by bolts, and a groove is opened on the top outer wall of the fixed block 28, and the first drive gear 24 and the second drive gear 25 are respectively connected to the bottom of the groove by guide shafts, and the first drive gear 24 and the second drive gear 25 are meshed with each other, and the linkage gear 23 is connected to the top of the guide shaft in the first drive gear 24 by a thread, and the linkage gear 23 is meshed with the secondary gear 22, and the cam 26 is fixed to the bottom end of the guide shaft of the second drive gear 25 by a thread, and the turntable 27 is rotatably connected to the U-shaped bracket 14 near the bottom outer wall of the fixed block 28 by a connecting rod, and the top outer wall of the turntable 27 is provided with meshing grooves distributed at equal distances, and the cam 2 6 is adapted to the meshing groove, the bottom outer wall of the turntable 27 is connected to the cross guide 6 through a fixing rod, and the surrounding structure of the cross guide 6 is inclined, and the camera 7 is fixed to the bottom outer wall of the four sides of the cross guide 6 by bolts; when working, the rotating second gear 22 drives the linkage gear 23 to rotate, and the first driving gear 24 is rotated through the linkage gear 23. At the same time, through the relationship of meshing with the second driving gear 25, the cam 26 at the bottom of the second driving gear 25 is driven to rotate, and the cam 26 is adapted to the meshing groove, thereby driving the turntable 27 to rotate intermittently, and the intermittently rotating turntable 27 effectively drives the camera 7 of the cross guide 6 at its bottom to rotate intermittently, which not only ensures that the camera 7 can clearly collect human feature images during the intermittent process, but also ensures multi-angle shooting of human weight recognition images, thereby promoting recognition accuracy.

[0073] A winding drum 19 is fixed to the outer wall of the main frame 1 near the top of the guide groove 3 by bolts, and an elastic sealing plate is connected to the winding drum 19 by a ruler spring, and a handle 20 is fixed to one end of the elastic sealing plate by bolts, and clamping protrusions are provided on both sides of the end of the elastic sealing plate, and a clamping seat is fixed to the bottom side of the main frame 1 by bolts, and the clamping protrusions are adapted to the clamping seat; by pulling out the elastic sealing plate in the winding drum 19, the safety of the laser contour sensor 17 is effectively guaranteed to avoid dust falling on its surface and reducing its function.

[0074] When this embodiment is in use, the person to be identified passes through the two main frames 1 and stands between the two main frames 1. At this time, the motor 15 is started, and the motor 15 drives the second main gear 21 to rotate, and promotes the rotation of the rotating rod 13 through the secondary gear 22. At the same time, the rotating rotating rod 13 promotes the rotation of the secondary gears 9 at both ends thereof, thereby driving the two lead screws 10 in the main frame 1 to rotate, and the rotating lead screw 10 promotes the fixed plate 11 and its two guide plates 12 to move up and down in the slide groove 18. In the meantime, the laser profile sensor 17 on one side of the fixed plate 11 is started, and the lifting laser profile sensor 17 adopts the laser triangulation reflection principle. The laser beam is amplified to form a laser line and projected onto the surface of the human body. The reflected light passes through the high-quality optical The system is projected onto the imaging matrix, and the distance from the sensor to the measured surface and the position information along the laser line are obtained through calculation. The three-dimensional contour of the measured target can be directly obtained through data processing. At the same time, the rotating second gear 22 drives the linkage gear 23 to rotate, and the first drive gear 24 is rotated through the linkage gear 23. At the same time, through the engagement with the second drive gear 25, the cam 26 at the bottom of the second drive gear 25 is driven to rotate, and the cam 26 is adapted to the engagement groove, thereby driving the turntable 27 to rotate intermittently, and the intermittently rotating turntable 27 effectively drives the camera 7 of the cross guide 6 at its bottom to rotate intermittently, thereby completing the feature image acquisition of the human body and promoting the effective acquisition and uploading of the feature image of the human body.

[0075] Example 3:

[0076] A human weight recognition system based on local features, such as Figure 6-8 As shown, a mounting housing 29 is fixed to an outer wall of one side of the main frame 1 by bolts, and a start-stop mechanism is provided inside the mounting housing 29 .

[0077] The start-stop mechanism includes a pressing buckle 30, a liquid outlet pipe 31, a liquid inlet pipe 32, a pressing valve 33, a pedal 34, a hydraulic rotating assembly 37, a turntable 38, a limit clamp 39, a block 40, an inclined hole 41, a spring 42 and a inclined top 43. The pressing buckle 30 is slidably connected to the outer wall on the top side of the mounting casing 29, and the liquid outlet pipe 31 and the liquid inlet pipe 32 are respectively fixed to the outer wall on the front side of the mounting casing 29 by more than three clamps. The pressing valve 33 is fixed to the bottom side of the mounting casing 29 by bolts, and the port of the pressing valve 33 is connected to the port of the liquid inlet pipe 32 by threads. One end of the pedal 34 is fixed to the pressing end of the pressing valve 33 by a torsion spring, and counterweights 36 are fixed on both sides of the end of the elastic sealing plate by bolts.

[0078] The main frame 1 is fixed with two magnetic seats 35 by bolts on one side near the press valve 33. The magnetic seat 35 is adapted to the counterweight 36. The inclined top 43 is welded to the top of the press buckle 30, and a guide groove is opened on the side of the mounting housing 29 near the upper side of the inclined top. The spring 42 is welded to the inner wall of one side of the guide groove. The clamping block 40 is horizontally slidably connected to the two inner sides of the guide groove, and one end of the clamping block 40 and one end of the spring 42 are welded to each other. The inclined hole 41 is opened at the top of the clamping block 40, and the inclined top 43 is adapted to the inclined hole 41. The inner walls on both sides of the winding drum 19 are connected by bearings for rotation. It is connected to a winding roller, and the turntable 38 is connected to one end of the winding roller through a connecting shaft. More than three limit clamps 39 are fixed to the circumferential outer wall of the turntable 38 at equal distances by bolts. A limit groove is opened on one side of each limit clamp 39, and the limit groove is adapted to the end of the clamping block 40. The hydraulic rotating component 37 is connected to the connecting shaft end of the inner ring of the turntable 38 through a connecting shaft. The two sides of the hydraulic rotating component 37 are respectively connected to the top of the liquid outlet pipe 31 and the liquid inlet pipe 32 through threads, and the bottom ports of the liquid outlet pipe 31 and the liquid inlet pipe 32 are connected to the hydraulic system.

[0079] When the entire device is not in use, the pressing buckle 30 is squeezed upward, so that the inclined top 43 at the top of the pressing buckle 30 is displaced upward, and the displaced inclined top 43 passes through the inclined hole 41 in the block 40, so that the block 40 slides horizontally in the guide groove. At this time, the spring 42 on one side of the block 40 is tightened, and one end of the block 40 is disengaged from the limit clamping plate 39 on the circumference of the turntable 38. At the same time, the elastic sealing plate wrapped around the winding roller is pulled out from the winding drum 19 under the action of the two counterweights 36 at its end, and moves vertically downward. When the end of the elastic sealing plate falls to the bottom of the main frame 1, the two magnetic seats 35 adsorb the two counterweights 36, thereby completing the sealing effect of the elastic sealing plate on the inside of the device, while reducing the workload of the staff, and promoting the protection of the laser profile sensor 17 and its components.

[0080] A copper block 44 is bonded to the outer wall of the top of the counterweight block 44, and a power-on cavity is provided on the winding drum 19 near the top of the copper block 44. Two connection paddles 45 are provided on the outer wall of the power-on cavity near the top of the copper block 44, and a power-on wire 46 is fixed to the inner wall of one side of the power-on cavity by bolts, and the connection paddle 45 is welded to the power-on end of the power-on wire 46.

[0081] When the device is activated, the pedal 34 is pressed by the foot, causing one end of the pedal 34 to tilt. At this time, the tilted end of the pedal 34 triggers the pressing valve 33, causing the hydraulic system to work and introduce hydraulic oil into the liquid inlet pipe 32. The hydraulic oil enters the hydraulic rotating assembly 37 through the liquid inlet pipe 32, so that the connecting shaft at one end drives the turntable 38 to rotate. At this time, the rotating turntable 38 drives the winding roller to rotate through the connecting shaft, thereby causing the elastic sealing plate outside the winding drum 19 to rise upward and perform the winding work. When the winding work is completed, the counterweight block 36 on one side of the end of the elastic sealing plate sends the copper block 44 into the electrified cavity, and the copper block 44 now connects the two connecting paddles 45 in the electrified cavity, so that electricity is supplied to the connecting wire 36, thereby driving the motor 15 in the chassis 5 to start working. Therefore, while lifting the elastic sealing plate, the startup action of the entire device is simplified and automation is achieved.

[0082] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A human weight recognition system based on local features, including an information acquisition module, a cloud database, an image comparison and recognition module, an object recognition log, and a security management module, characterized in that: The information collection module includes biometric feature collection, voice collection, and behavioral feature collection, and the cloud database includes data analysis and data management. The image comparison and recognition module includes image acceptance, preprocessing, feature extraction, feature selection, and analysis and comparison judgment. The object recognition log includes log storage and log management. The security management module includes recognition warning and information security encryption. The information collection module includes the following steps: A1: The person to be identified enters the collection door and stands inside the collection door; A2: The acquisition door uses cameras to perform temperature spectrum recognition, facial recognition, iris recognition, body feature recognition, voice feature recognition, and fingerprint recognition on the person to be identified to collect relevant image and text data; A3: Next, the collected data is archived and the collected information is transferred to the cloud database for analysis and comparison; The image comparison and recognition module includes the following steps: B1: Image acceptance: accept the image information of the biometric features of the person to be identified and save it for archiving; B2: Preprocessing: Encode and compress image information to reduce the amount of image data, enhance and restore the image to make the feature contours and details clear, and use filtering methods to ensure the image details; B3: Feature extraction: Extract multiple local features from the processed image, divide and store each local feature, and send it to the cloud database for feature comparison and analysis; B4: Feature selection: The cloud database will feed back the comparison results obtained through analysis, and select and extract the local features of the target in the archive; B5: Analysis, comparison and discrimination: Perform feature analysis on the selected target local features and compare them with the local features in the cloud database to obtain the discrimination results; The object identification log includes the following steps: C1: Log storage: The daily human weight recognition information is archived and integrated, uploaded to the cloud database, and each sub-information database is filled daily; C2: Log Management: Integrate the characteristic information in each sub-information database, remove the same characteristic information and supplement the relevant characteristic information, and store the sub-information database securely. The security management module includes the following steps: D1: Identification and warning: The feedback information is marked in red, saved and archived, and the relevant feature information is compared and identified; D2: Information security encryption: Perform security encryption on the sub-information database and mark it as a distinguishing mark; The collecting door comprises two main frames (1) and a top plate (4), and the outer walls of the two main frames (1) on opposite sides are respectively provided with guide grooves (3), and both sides of the two guide grooves (3) are provided with slide grooves (18), and the outer wall of the main frame (1) on the side close to the guide groove (3) is provided with three warning lights (2), and the inner wall of the main frame (1) on the side close to the guide groove (3) is provided with a lead screw (10), and the circumferential outer wall of the lead screw (10) is sleeved with a fixed plate (11), and the outer walls on both sides of the fixed plate (11) are respectively provided with There is a guide plate (12), and the opposite sides of the two guide plates (12) are both in sliding cooperation with the slide groove (18). A laser profile sensor (17) is provided on the outer wall of one side of the fixed plate (11). The top ends of the two lead screws (10) are provided with a same gear transmission mechanism, and the middle bottom end of the gear transmission mechanism is fixed with an intermittent rotation mechanism through a U-shaped bracket (14). Cameras (7) are provided around the bottom of the intermittent rotation mechanism, and the intermittent rotation mechanism and the gear transmission mechanism are linked to each other. A machine box (5) is provided on the top middle outer wall of the top plate (4), and a limiting mirror (16) is provided on the outer wall of one side of the guide plate (12). The gear transmission mechanism includes two first main gears (8), two first gears (9), a rotating rod (13), a motor (15), a second main gear (21) and a second gear (22), and the motor (15) is provided inside the machine box (5). The second main gear (21) is connected to the output end of the motor (15) through a connecting shaft. The rotating rod (13) is fixed to the outer wall of the motor (15) through a bearing. On the inner walls of both sides of the top plate (4), the secondary gear (22) is sleeved on the middle circumference of the rotating rod (13), and the second main gear (21) and the secondary gear (22) are meshed with each other. The two first main gears (8) are respectively arranged on the top circumferences of the two lead screws (10). The two first gears (9) are respectively sleeved on the two ends of the rotating rod (13). The first main gear (8) and the first gear (9) are meshed with each other. The U-shaped bracket (14) is connected to the circumference of the rotating rod (13) through a bearing.

2. A human weight recognition system based on local features according to claim 1, characterized in that: The intermittent rotation mechanism comprises a linkage gear (23), a first drive gear (24), a second drive gear (25), a cam (26), a turntable (27) and a fixed block (28), wherein the fixed block (28) is arranged on the middle outer wall of the bottom end of the U-shaped bracket (14), and a groove is formed on the top outer wall of the fixed block (28). The first drive gear (24) and the second drive gear (25) are respectively connected to the bottom of the groove by a guide shaft, and the first drive gear (24) and the second drive gear (25) are meshed with each other, and the linkage gear (23) is arranged at the top end of the guide shaft in the first driving gear (24), the linkage gear (23) and the secondary gear (22) are meshed with each other, the cam (26) is arranged at the bottom end of the guide shaft of the second driving gear (25), and the turntable (27) is rotatably connected to the bottom outer wall of the U-shaped bracket (14) close to the fixed block (28) through a connecting rod, the top outer wall of the turntable (27) is provided with meshing grooves distributed at equal distances, and the cam (26) and the meshing grooves are adapted to each other, and the bottom outer wall of the turntable (27) is connected to the cross guide frame (6) through a fixed rod.

3. A human weight recognition system based on local features according to claim 2, characterized in that: A winding drum (19) is provided on the outer wall of one side of the main frame (1) near the top of the guide groove (3), and an elastic sealing plate is connected to the winding drum (19) through a ruler spring, a handle (20) is provided at one end of the elastic sealing plate, and clamping protrusions are provided on both sides of the end of the elastic sealing plate, and a clamping seat is provided on one side of the bottom of the main frame (1), and the clamping protrusion is adapted to the clamping seat.

4. The human weight recognition system based on local features according to claim 3, characterized in that: A mounting casing (29) is provided on one side outer wall of the main frame (1), and a start-stop mechanism is provided inside the mounting casing (29).

Images