Portable three-dimensional imaging security check system capable of automatic detection
By combining terahertz imaging and deep learning algorithms, the portable three-dimensional imaging security inspection system solves the portability and intelligence problems of existing security inspection equipment, realizes fast and accurate dangerous goods detection, and is suitable for security inspection needs in high-density human flow environments.
Patent Information
- Application Number
- CN202510594381.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-09
- Publication Date
- 2025-09-05
AI Technical Summary
Existing security inspection equipment has the problems of low inspection efficiency, bulky and non-portable equipment, single function and insufficient intelligence. It is difficult to meet the needs of fast, accurate and intelligent security inspection in modern high-density human flow environments, especially when facing new types of dangerous goods.
A foldable mechanical platform and an intelligent detection platform are used, combined with terahertz imaging technology and deep learning algorithms to achieve three-dimensional imaging and automatic identification of samples to be detected. Three-dimensional images are constructed through terahertz radiation emission and detectors, and feature extraction and target detection are combined with deep convolutional neural networks to achieve real-time identification and graded warning of dangerous goods.
It has realized the intelligence and portability of security inspection equipment, improved the comprehensiveness and reliability of detection, significantly increased the detection speed and recognition accuracy, reduced the difficulty of operation, and is suitable for the security inspection needs of temporary public activities.
Smart Images
Figure CN120595388A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to security inspection equipment, in particular to a portable three-dimensional imaging security inspection system capable of automatic detection. Background Art
[0002] With rapid economic development and the continuous advancement of urbanization, significant progress has been made in the development of public transportation infrastructure. Transportation hubs such as high-speed rail, subways, and airports are becoming increasingly well-developed. Large-scale public events such as exhibitions, sporting events, and theatrical performances are being held more frequently, and the number of participants has exploded. However, large crowds also bring increased security risks. The probability of sudden public safety incidents in public places has significantly increased. Against this backdrop, security measures, as the first line of defense in maintaining public safety, have become increasingly important. Security checks can effectively intercept prohibited items such as controlled knives, flammable and explosive items, and dangerous chemicals, preventing potential threats, curbing dangerous activities, and protecting the lives and property of the people. Furthermore, strict security inspection systems can deter criminals and maintain public order and stability.
[0003] Currently, the security inspection methods commonly used in various public places mainly include traditional security inspection equipment such as X-ray security scanners, manual inspections, handheld metal detectors, metal detection gates, explosive trace detectors, and liquid detectors. These devices have obvious limitations in practical applications, as follows:
[0004] First, large equipment such as X-ray scanners and metal detectors are bulky and complex in structure, typically requiring fixed installation in specific locations. This makes transportation and deployment difficult, and makes it difficult to quickly set up security screening channels for temporary large-scale events.
[0005] Secondly, portable devices such as handheld metal detectors, explosives trace detectors, and liquid detectors have limited functionality and can only detect specific types of dangerous goods. They are inadequate to meet the increasingly diverse security threats, and their detection speeds are insufficient to meet the demands of high-density traffic flow. While manual inspections have high accuracy rates, they are inefficient and can easily cause congestion during peak hours, posing serious safety risks. More critically, these traditional security inspection devices rely heavily on manual operation and judgment. Security inspectors must focus their attention for extended periods observing X-ray images and rely on their personal experience to identify various types of dangerous goods. This is not only prone to misjudgments due to fatigue, but also suffers from significant experience-based and subjective variability. Studies have shown that after four hours of continuous work, security inspectors' accuracy in identifying dangerous goods can drop by over 30%. Furthermore, the emergence of new types of dangerous goods, such as 3D-printed weapons and nano-explosives, poses a significant challenge to the professional capabilities of security inspectors.
[0006] In summary, existing security inspection equipment suffers from widespread issues such as low inspection efficiency, bulky and unportable equipment, limited functionality, and insufficient intelligence. Furthermore, it is particularly inadequate when it comes to detecting new types of dangerous goods. These shortcomings severely restrict the practical application of security inspection systems at large-scale public events, making it difficult to meet the demands for rapid, accurate, and intelligent security inspections in modern, high-density environments.
[0007] Therefore, how to make security inspection intelligent and portable, improve the comprehensiveness and reliability of security inspection, and reduce the difficulty of operation is a technical problem that needs to be solved urgently. Summary of the Invention
[0008] The technical task of the present invention is to provide a portable three-dimensional imaging security inspection system with automatic detection to solve the problem of how to realize the intelligence and portability of security inspection, improve the comprehensiveness and reliability of security inspection, and reduce the difficulty of operation.
[0009] The technical task of the present invention is achieved in the following way: a portable three-dimensional imaging security inspection system capable of automatic detection, the system comprising a foldable mechanical platform and an intelligent detection platform, the foldable mechanical platform comprising a conveying platform, a left support column and a right support column are arranged on the conveying platform, a support beam is arranged between the left support column and the right support column, the two ends of the support beam are hingedly connected to the left support column and the right support column respectively, a terahertz reflector is arranged at one end of the support beam close to the left support column, a terahertz radiation emission source is arranged at the middle position of the right support column, a terahertz linear detector is arranged at one end of the conveying platform close to the left support column, and a terahertz linear detector is arranged at the upper position of the conveying platform close to one end of the right support column and Passive terahertz radiation detectors are installed at one end of the left support column close to the conveying platform. The terahertz radiation emission source emits terahertz radiation in a specific frequency band. When the terahertz radiation in the corresponding frequency band penetrates the sample to be detected, different substances show unique absorption and scattering characteristics. The terahertz linear detector receives the radiation signal after penetration. The passive terahertz radiation detector assists in imaging and converts the radiation signal after penetration into three-dimensional image information. The three-dimensional image information is transmitted to the intelligent detection platform. The intelligent detection platform integrates automatic recognition algorithms, uses super-resolution reconstruction technology to improve imaging quality, combines deep convolutional neural networks for feature extraction, and combines target detection algorithms to achieve real-time identification of dangerous goods.
[0010] Preferably, the intelligent detection platform includes:
[0011] The preprocessing module is used to perform adaptive mosaic enhancement, multi-scale hybrid enhancement, and GAN-based super-resolution reconstruction on 3D image information to improve the quality of the original terahertz image;
[0012] The feature extraction module is used to extract multi-scale features using an improved CSPDarknet deep convolutional neural network. The improved CSPDarknet deep convolutional neural network can simultaneously capture the global semantic features and local detail features of the target by introducing an attention mechanism and a feature pyramid structure.
[0013] The recognition module is used to achieve end-to-end dangerous goods recognition based on the improved YOLOv5 object detection framework; the improved YOLOv5 object detection framework introduces an adaptive anchor box mechanism and a SIoU loss function.
[0014] As a preferred option, the working process of the intelligent detection platform is as follows:
[0015] ① The preprocessing module performs preprocessing operations on the original terahertz image through adaptive noise reduction, contrast enhancement and super-resolution reconstruction technology to obtain the preprocessed image and improve the image quality;
[0016] ② The feature extraction module performs multi-scale feature extraction on the preprocessed image, capturing both the global semantic information and local detail features of the target;
[0017] ③ The target detection algorithm optimized based on transfer learning performs intelligent analysis on the extracted features, accurately identifies the category of dangerous goods through feature comparison of the recognition module, and automatically generates high-precision target detection frames;
[0018] ④ Implement graded warnings based on the identification results. Specifically, high-risk items such as explosives will trigger an audio-visual linkage alarm, medium-risk items such as controlled knives will trigger a visual warning, and low-risk items such as suspicious liquids will be pushed to the manual review queue to achieve intelligent graded disposal.
[0019] Preferably, the system also includes a shell, which covers the foldable mechanical platform and encloses the terahertz emission source, the terahertz linear detector and the terahertz reflector to form a terahertz light path channel for security inspection; at the same time, the shell separates the inspected object from the external environment, reducing the interference of electromagnetic radiation in the external environment on the detection results.
[0020] As a priority, the intelligent detection platform builds a terahertz hazardous materials database containing more than 10,000 annotated images in 12 categories, covering metal weapons, powder and liquid hazardous materials.
[0021] Preferably, the support beam is made of aviation aluminum alloy.
[0022] Preferably, the conveying platform adopts a double-belt conveyor with stepless speed regulation of 0.1-2.0m / s.
[0023] Preferably, the working process of the system is as follows:
[0024] S1. Unfold the foldable mechanical platform and install the terahertz emission source, terahertz linear detector, passive terahertz radiation detector and terahertz reflector;
[0025] S2, start the terahertz emission source, terahertz linear detector, passive terahertz radiation detector and intelligent detection platform;
[0026] S3. The luggage to be inspected is placed into the input port and transported out of the output port via the conveyor platform. During the transportation process, the terahertz wave source emits terahertz radiation, which penetrates the sample to be inspected and is received by the terahertz linear detector. The passive terahertz radiation detector receives the terahertz radiation spontaneously emitted in the horizontal direction from the item, and together they construct three-dimensional image information.
[0027] S4. The intelligent detection platform automatically detects the type and location of dangerous goods and issues an alarm. Specifically, the intelligent detection platform generates an original image, performs image enhancement on the input image through the preprocessing module, and then inputs it into the improved CSPDarknet deep convolutional neural network to extract features. The improved YOLOv5 target detection framework then uses the extracted features to identify the possible category of dangerous goods and automatically calibrates the target frame. If it is a dangerous good, different sound and light signals are used to alarm according to the category of the dangerous good. Otherwise, the next image is cycled.
[0028] S5. Unpack suspected dangerous goods for inspection: When prohibited items are detected in the luggage, the items need to be carefully inspected.
[0029] The portable three-dimensional imaging security inspection system capable of automatic detection of the present invention has the following advantages:
[0030] (1) This invention achieves breakthroughs in the intelligence and portability of security inspection equipment by integrating multiple core technologies such as terahertz imaging technology, deep learning algorithms, and modular mechanical design;
[0031] (2) The present invention significantly reduces the size and weight of the device through a miniaturized, modular, and detachable terahertz wave emission and intelligent detection platform and a foldable mechanical platform, significantly improving portability and rapid deployment capabilities;
[0032] (3) Based on advanced terahertz imaging technology, this invention achieves high-precision detection of dangerous goods made of various materials, such as ceramic products, plastic containers, and powdered substances, significantly improving the comprehensiveness and reliability of detection;
[0033] (4) This invention innovatively uses a deep learning target detection algorithm to replace traditional matching methods, significantly improving key indicators such as detection speed, recognition accuracy, and false alarm rate;
[0034] (5) The present invention adopts a modular architecture and networked design, supports flexible expansion of functional components and algorithm upgrades, and users can perform personalized configuration according to actual needs;
[0035] (6) The optimized human-computer interaction interface of the present invention provides one-touch intelligent operation and intuitive visual detection results, greatly reducing the difficulty of operation;
[0036] (7) The present invention is particularly suitable for security inspection needs in special scenarios such as temporary public activities. It is superior to similar products in terms of detection efficiency, accuracy and ease of use. It provides an effective technical solution for improving the security prevention capabilities of public places and has significant market application value.
[0037] Therefore, the present invention has the characteristics of reasonable design, simple structure, easy processing, small size, convenient use, and multiple uses, and thus has good promotion and use value. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] The present invention will be further described below with reference to the accompanying drawings.
[0039] Attachment Figure 1 The figure is a schematic diagram of the structure of a portable three-dimensional imaging security inspection system capable of automatic detection;
[0040] Attachment Figure 2 A schematic diagram of the foldable mechanical platform in an unfolded state;
[0041] Attachment Figure 3 is a schematic diagram of the foldable mechanical platform in a folded state;
[0042] Attachment Figure 4 This is a flowchart of the working process of the intelligent detection platform;
[0043] Attachment Figure 5 Schematic diagram of the network structure of the YOLOv5 target detection framework. DETAILED DESCRIPTION
[0044] The portable three-dimensional imaging security inspection system capable of automatic detection of the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.
[0045] In the description of the present invention, it should be understood that terms such as "upper," "lower," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer" are used to indicate positions or relationships based on the positions or relationships shown in the accompanying drawings and are intended solely to facilitate description and simplify the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and are therefore not to be construed as limitations of the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
[0046] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.
[0047] Example:
[0048] As attached Figure 1 As shown, this embodiment provides a portable three-dimensional imaging security inspection system that can automatically detect, and its structure includes a foldable mechanical platform and an intelligent detection platform. The foldable mechanical platform includes a conveying platform, and a left support column and a right support column are installed on the conveying platform. A support beam is installed between the left support column and the right support column, and the two ends of the support beam are hingedly connected to the left support column and the right support column respectively. A terahertz reflector is installed at one end of the support beam close to the left support column, and a terahertz radiation emission source is installed in the middle position of the right support column. A terahertz linear detector is installed at one end of the conveying platform close to the left support column, and the upper position of the conveying platform close to one end of the right support column and the left support column are provided. Passive terahertz radiation detectors are installed at one end of the column close to the conveying platform. The terahertz radiation emission source emits terahertz radiation in a specific frequency band. When the terahertz radiation in the corresponding frequency band penetrates the sample to be detected, different substances show unique absorption and scattering characteristics. The terahertz linear detector receives the radiation signal after penetration. The passive terahertz radiation detector assists in imaging and converts the radiation signal after penetration into three-dimensional image information. The three-dimensional image information is transmitted to the intelligent detection platform. The intelligent detection platform integrates automatic recognition algorithms, uses super-resolution reconstruction technology to improve imaging quality, combines deep convolutional neural networks for feature extraction, and combines target detection algorithms to achieve real-time recognition of dangerous goods.
[0049] In terms of imaging, this embodiment uses active transmissive terahertz imaging as the primary imaging mode, combined with passive terahertz imaging as an auxiliary imaging method to form a complementary detection mechanism. When this embodiment is in operation, the object to be inspected is placed in the detection area between the terahertz radiation emission source and the terahertz linear detector. A precision mechanical scanning system is used to achieve three-dimensional relative movement between the terahertz radiation emission source, the terahertz linear detector, and the object to be inspected. Based on the specific absorption and scattering characteristics of different substances for terahertz waves, the characteristic spectrum information of various types of dangerous goods can be accurately obtained. For example, metal objects will produce obvious reflection signals, ceramic materials will exhibit specific absorption peaks, and powdered explosives and drugs will exhibit unique terahertz fingerprint characteristics. In terms of intelligent detection, intelligent detection integrates an advanced automatic recognition algorithm platform, uses super-resolution image reconstruction technology to improve imaging quality, combines deep convolutional neural networks for feature extraction, and uses target detection algorithms to achieve real-time identification of dangerous goods.
[0050] The intelligent scoring detection platform in this embodiment includes:
[0051] The preprocessing module is used to perform adaptive mosaic enhancement, multi-scale hybrid enhancement, and GAN-based super-resolution reconstruction on 3D image information to improve the quality of the original terahertz image;
[0052] The feature extraction module is used to extract multi-scale features using an improved CSPDarknet deep convolutional neural network. The improved CSPDarknet deep convolutional neural network can simultaneously capture the global semantic features and local detail features of the target by introducing an attention mechanism and a feature pyramid structure.
[0053] The recognition module is used to achieve end-to-end dangerous goods recognition based on the improved YOLOv5 object detection framework; the improved YOLOv5 object detection framework introduces an adaptive anchor box mechanism and a SIoU loss function.
[0054] The intelligent detection platform in this embodiment builds a terahertz hazardous materials database containing more than 10,000 annotated images of 12 categories, covering metal weapons, powder and liquid hazardous materials.
[0055] As attached Figure 4 The working process of the intelligent detection platform in this embodiment is as follows:
[0056] ① The preprocessing module performs preprocessing operations on the original terahertz image through adaptive noise reduction, contrast enhancement and super-resolution reconstruction technology to obtain the preprocessed image and improve the image quality;
[0057] ② The feature extraction module performs multi-scale feature extraction on the preprocessed image, capturing both the global semantic information and local detail features of the target;
[0058] ③ The target detection algorithm optimized based on transfer learning performs intelligent analysis on the extracted features, accurately identifies the category of dangerous goods through feature comparison of the recognition module, and automatically generates high-precision target detection frames;
[0059] ④ Implement graded warnings based on the identification results. Specifically, high-risk items such as explosives will trigger an audio-visual linkage alarm, medium-risk items such as controlled knives will trigger a visual warning, and low-risk items such as suspicious liquids will be pushed to the manual review queue to achieve intelligent graded disposal.
[0060] The foldable mechanical platform in this embodiment adopts an innovative modular architecture, which significantly improves the rapid deployment capability and ease of use of the equipment through multiple structural optimizations. The support beam is made of aviation aluminum alloy and is processed with precision CNC to ensure that the structure is lightweight while maintaining sufficient mechanical strength and stability. The conveying platform can achieve dual-belt conveying with stepless speed regulation of 0.1-2.0m / s. Figure 2 As shown, the support beams can be folded to compress the volume when not in use, and leave a detection channel and an optical path channel when in use, so that the volume is smaller in the non-working state, which is different from carrying and transportation.
[0061] This embodiment also includes a shell, which covers the foldable mechanical platform and encloses the terahertz emission source, the terahertz linear detector and the terahertz reflector, forming a terahertz light path channel for security inspection; at the same time, the shell separates the inspected object from the external environment, reducing the interference of electromagnetic radiation in the external environment on the detection results.
[0062] This embodiment utilizes advanced terahertz imaging detection technology. Its operating principle is as follows: A high-precision terahertz source emits terahertz radiation in a specific frequency band. When this radiation penetrates the sample to be detected, different materials exhibit unique absorption and scattering characteristics. A highly sensitive terahertz linear detector array then receives the transmitted radiation signal, which is then imaged and converted into three-dimensional image information using passive terahertz radiation detectors. After being transmitted to the intelligent detection platform, this raw image data undergoes image preprocessing, including noise reduction, contrast optimization, and super-resolution reconstruction, significantly improving image quality. Next, an improved deep convolutional neural network extracts multi-level features from the preprocessed image. Combined with an optimized object detection algorithm, this feature is intelligently analyzed to accurately identify and classify various types of hazardous materials. The system ultimately automatically generates precise bounding boxes for each detected hazardous object and triggers corresponding alarm mechanisms based on the hazard level, completing a fully automated hazardous material identification process from detection to early warning. The entire detection process is completed in milliseconds, achieving rapid and accurate security inspections.
[0063] The working process of this embodiment is as follows:
[0064] S1. Unfold the foldable mechanical platform and install the terahertz emission source, terahertz linear detector, passive terahertz radiation detector and terahertz reflector; as shown in the attached Figure 2 As shown in the figure, fold the left support column, support beam and right support column in the direction of the arrow to obtain the attached Figure 3 shown in the folded state;
[0065] S2, start the terahertz emission source, terahertz linear detector, passive terahertz radiation detector and intelligent detection platform;
[0066] S3. Place the baggage to be inspected at input port A and transport it out of output port B via the conveyor platform. During transportation, the terahertz wave source emits terahertz radiation, which penetrates the sample to be inspected and is received by the terahertz linear detector. The passive terahertz radiation detector receives the terahertz radiation spontaneously emitted horizontally from the item, and together they construct a three-dimensional image.
[0067] S4. The intelligent detection platform automatically detects the type and location of dangerous goods and issues an alarm. Specifically, the intelligent detection platform generates an original image, performs image enhancement on the input image through the preprocessing module, and then inputs it into the improved CSPDarknet deep convolutional neural network to extract features, and then uses the improved YOLOv5 target detection framework (see attached) to extract features. Figure 5 (As shown) Based on the extracted features, the possible dangerous goods category is determined and the target frame is automatically calibrated: if it is a dangerous good, different sound and light signal alarms are implemented according to the dangerous goods category, otherwise the next image is cycled;
[0068] S5. Unpack suspected dangerous goods for inspection: When prohibited items are detected in the luggage, the items need to be carefully inspected.
[0069] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A portable three-dimensional imaging security inspection system capable of automatic detection, characterized in that: The system includes a foldable mechanical platform and an intelligent detection platform. The foldable mechanical platform includes a conveying platform. A left support column and a right support column are provided on the conveying platform. A support beam is provided between the left support column and the right support column. The two ends of the support beam are respectively hingedly connected to the left support column and the right support column. A terahertz reflector is provided at one end of the support beam close to the left support column, a terahertz radiation emission source is provided at the middle position of the right support column, a terahertz linear detector is provided at one end of the conveying platform close to the left support column, and a terahertz linear detector is provided at the upper position of the conveying platform close to one end of the right support column and the end of the left support column close to the conveying platform. There are passive terahertz radiation detectors. The terahertz radiation emission source emits terahertz radiation in a specific frequency band. When the terahertz radiation in the corresponding frequency band penetrates the sample to be detected, different substances show unique absorption and scattering characteristics. The terahertz linear detector receives the radiation signal after penetration. The passive terahertz radiation detector assists in imaging and converts the radiation signal after penetration into three-dimensional image information. The three-dimensional image information is transmitted to the intelligent detection platform. The intelligent detection platform integrates automatic recognition algorithms, uses super-resolution reconstruction technology to improve imaging quality, combines deep convolutional neural networks for feature extraction, and combines target detection algorithms to achieve real-time identification of dangerous goods.
2. The portable three-dimensional imaging security inspection system capable of automatic detection according to claim 1 is characterized in that: The intelligent detection platform includes: The preprocessing module is used to perform adaptive mosaic enhancement, multi-scale hybrid enhancement, and GAN-based super-resolution reconstruction on 3D image information to improve the quality of the original terahertz image; The feature extraction module is used to extract multi-scale features using an improved CSPDarknet deep convolutional neural network. The improved CSPDarknet deep convolutional neural network can simultaneously capture the global semantic features and local detail features of the target by introducing an attention mechanism and a feature pyramid structure. The recognition module is used to achieve end-to-end dangerous goods recognition based on the improved YOLOv5 object detection framework; the improved YOLOv5 object detection framework introduces an adaptive anchor box mechanism and a SIoU loss function.
3. The portable three-dimensional imaging security inspection system capable of automatic detection according to claim 1 is characterized in that: The working process of the intelligent detection platform is as follows: ① The preprocessing module performs preprocessing operations on the original terahertz image through adaptive noise reduction, contrast enhancement and super-resolution reconstruction technology to obtain the preprocessed image and improve the image quality; ② The feature extraction module performs multi-scale feature extraction on the preprocessed image, capturing both the global semantic information and local detail features of the target; ③ The target detection algorithm optimized based on transfer learning performs intelligent analysis on the extracted features, accurately identifies the category of dangerous goods through feature comparison of the recognition module, and automatically generates high-precision target detection frames; ④ Implement graded warnings based on the identification results. Specifically, high-risk items will trigger an audio-visual linkage alarm, medium-risk items will trigger a visual warning, and low-risk items will be pushed to the manual review queue to achieve intelligent graded disposal.
4. The portable three-dimensional imaging security inspection system capable of automatic detection according to claim 1, characterized in that: The system also includes a shell, which covers the foldable mechanical platform and encloses the terahertz emission source, terahertz linear detector and terahertz reflector to form a terahertz light path channel for security inspection; at the same time, the shell separates the inspected objects from the external environment, reducing the interference of electromagnetic radiation in the external environment on the detection results.
5. The portable three-dimensional imaging security inspection system capable of automatic detection according to claim 1 is characterized in that: The intelligent detection platform builds a terahertz hazardous materials database with annotated images, covering metal weapons, powder and liquid hazardous materials.
6. The portable three-dimensional imaging security inspection system capable of automatic detection according to claim 1, characterized in that: The support beam is made of aviation aluminum alloy.
7. The portable three-dimensional imaging security inspection system capable of automatic detection according to claim 1 is characterized in that: The conveying platform adopts double belt conveyor with stepless speed regulation of 0.1-2.0m / s.
8. The portable three-dimensional imaging security inspection system capable of automatic detection according to any one of claims 1 to 7, characterized in that: The working process of the system is as follows: S1. Unfold the foldable mechanical platform and install the terahertz emission source, terahertz linear detector, passive terahertz radiation detector and terahertz reflector; S2, start the terahertz emission source, terahertz linear detector, passive terahertz radiation detector and intelligent detection platform; S3. The luggage to be inspected is placed into the input port and transported out of the output port via the conveyor platform. During transportation, the terahertz wave source emits terahertz radiation, which penetrates the sample to be tested and is received by the terahertz linear detector. The passive terahertz radiation detector receives the spontaneous terahertz radiation from the horizontal direction of the object, and together they construct three-dimensional image information. S4. The intelligent detection platform automatically detects the type and location of dangerous goods and issues an alarm. Specifically, the intelligent detection platform generates an original image, performs image enhancement on the input image through the preprocessing module, and then inputs it into the improved CSPDarknet deep convolutional neural network to extract features. The improved YOLOv5 target detection framework then uses the extracted features to identify the possible category of dangerous goods and automatically calibrates the target frame. If it is a dangerous good, different sound and light signals are used to alarm according to the category of the dangerous good. Otherwise, the next image is cycled. S5. Unpack suspected dangerous goods for inspection: When prohibited items are detected in the luggage, the items need to be carefully inspected.