DRAWINGS
2 5 3 FIG. 1
11
FIG. 2
IMAGE TARGET RECOGNITION SYSTEM BASED ON RGB DEPTH-OF-FIELD CAMERA AND HYPERSPECTRAL CAMERA TECHNICAL FIELD The present disclosure relates to the technical field of image target recognition systems, and in particular, to an image target recognition system based on an RGB depth-of-field (DOF) camera and a hyperspectral camera. BACKGROUND An existing RGB camera obtains a two-dimensional image containing a target, and a hyperspectral camera scans row information of the two-dimensional image to obtain spectral information of the target, so as to recognize the target. However, there are following disadvantages: (1) If texture and other information of a target image needs to be obtained, the image target recognized by the RGB camera and the hyperspectral camera has low definition, and the texture and other information cannot be viewed. (2) Highlight degrees of all images of the image target quickly recognized by the RGB camera and the hyperspectral camera are consistent, so that the target cannot be quickly distinguished from other details around the target. Therefore, there is a need for an image target recognition system that can quickly recognize a target and highlight the target and its texture and other information. SUMMARY The present disclosure aims to provide an image target recognition system based on an RGB DOF camera and a hyperspectral camera, which solves the prior-art problem that when image target recognition is performed only through an RGB camera and a hyperspectral camera, an image target has low definition and is not highlighted, and image texture information cannot be obtained. The present disclosure can quickly and accurately recognize a target and provide a high-clarity recognized image. To achieve the above objectives, the present disclosure provides the following solutions: An image target recognition system includes: a hyperspectral camera, an RGB DOF camera, and a processor, where the processor is separately connected to the hyperspectral camera and the RGB DOF camera; the hyperspectral camera is configured to scan an image and obtain spectral information of the image; the processor is configured to match the spectral information of the image to obtain spectral information of a target in the image, and trigger the RGB DOF camera when the spectral information of the target in the image is obtained; and the RGB DOF camera is configured to collect three-dimensional (3D) image information of the target. Optionally, the image target recognition system further includes: an illumination device, where the illumination device is connected to the processor; the processor is configured to adjust illuminance of the illumination device; and the RGB DOF camera is configured to adjust an operating parameter based on the illuminance. Optionally, the image target recognition system further includes: a PC, where the PC is connected to the processor; and the PC is configured to display the target in the image. Optionally, the image target recognition system further includes: a casing, where the processor and the PC are both arranged inside the casing; the hyperspectral camera and the RGB DOF camera are arranged on an outer surface of the casing; and the illumination device is arranged on both sides of the casing. Optionally, the hyperspectral camera specifically includes: an imaging lens and an imaging unit, where the imaging unit is separately connected to the imaging lens and the processor. Optionally, the RGB DOF camera specifically includes: an infrared camera, a laser, and an RGB camera, where the infrared camera, the laser, and the RGB camera are all connected to a controller; and the laser is configured to obtain distances of different points in the image. Optionally, the processor is afield programmable gate array (FPGA). Optionally, there are two illumination devices; there are two imaging lenses; and there are two infrared cameras. Compared with the prior art, the present disclosure has the following beneficial effects: The present disclosure proposes an image target recognition system based on an RGB DOF camera and a hyperspectral camera. The RGB DOF camera is used to collect 3D image information of a target, which improves clarity of the image target, so that image information such as a highlight degree and texture can be obtained. This solves a prior-art problem that when image target recognition is performed only through an RGB camera and a hyperspectral camera, an image target has low definition and is not highlighted, and image texture information cannot be obtained. The present disclosure can quickly and accurately recognize a target and provide a high-clarity recognized image. According to the present disclosure, a high-clarity DOF image is obtained, a recognized target image is clear, and details such as image texture and coarseness can be obtained. Through analysis and processing on the DOF image, strong technical support can be provided for face recognition, foreign object detection, and the like. According to the present disclosure, the RGB DOF camera is triggered based on spectral information for timely collection, which can avoid a large number of data processing and data analysis processes. The present disclosure triggers the DOF camera based on the spectral information, which can avoid the shortcoming of untimely triggering due to harsh environments, and further improves real-time performance of image target recognition. In addition, the RGB DOF camera of the present disclosure is self-adaptive, and can adjust a corresponding operating parameter based on illuminance, to achieve accurate imaging under strong light or dark environments, provide better 3D DOF detail images, and further improve the clarity of an image target. BRIEF DESCRIPTION OF DRAWINGS To describe the technical solutions in the embodiments of the present disclosure or in the prior art or the prior art more clearly, the following briefly describes the accompanying drawings required for describing the embodiments. Apparently, the accompanying drawings in the following description show merely some embodiments of the present disclosure, and a person of ordinary skill in the art may still derive other drawings from these accompanying drawings without creative efforts. FIG. 1 is a schematic structural diagram of an image target recognition system according to an embodiment of the present disclosure. FIG. 2 is a front view of an image target recognition system according to an embodiment of the present disclosure. FIG. 3 is an effect diagram of an image target recognition system according to an embodiment of the present disclosure. Reference numerals: 1. light source; 2. imaging lens; 3. infrared camera B; 4. infrared camera A; 5. RGB camera; 6. laser; and 7. casing. DETAILED DESCRIPTION The technical solutions of the embodiments of the present disclosure are clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are merely a part rather than all of the embodiments of the present disclosure. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present disclosure without creative efforts shall fall within the protection scope of the present disclosure. The present disclosure aims to provide an image target recognition system based on an RGB DOF camera and a hyperspectral camera, which solves a prior-art problem that when image target recognition is performed only through an RGB camera and a hyperspectral camera, an image target has low definition and is not highlighted, and image texture information cannot be obtained. The present disclosure can quickly and accurately recognize a target and provide a high-clarity recognized image. To make the objectives, features, and advantages of the present disclosure more obvious and comprehensive, the following further describes in detail the present disclosure with reference to the accompanying drawing and specific implementations. Embodiment 1 FIG. 1 is a schematic structural diagram of an image target recognition system according to an embodiment of the present disclosure, and FIG. 2 is a front view of an image target recognition system according to an embodiment of the present disclosure. As shown in FIG. 1 and FIG. 2, an image target recognition system based on an RGB DOF camera and a hyperspectral camera includes: a hyperspectral camera (CCD camera), an RGB DOF camera, a processor (not shown in the figures), an illumination device, a PC, and a casing 7. The processor is separately connected to the hyperspectral camera and the RGB DOF camera; the hyperspectral camera is configured to scan an image and obtain spectral information of the image; the processor is configured to match the spectral information of the image to obtain spectral information of a target in the image, and trigger the RGB DOF camera when the spectral information of the target in the image is obtained; and the RGB DOF camera is configured to collect 3D image information of the target. The illumination device is connected to the processor; the processor is configured to adjust illuminance of the illumination device; and the RGB DOF camera is configured to adjust an operating parameter based on the illuminance. The PC is connected to the processor; and the PC is configured to display the target in the image. The processor and the PC are both arranged inside the casing; the hyperspectral camera and the RGB DOF camera are arranged on an outer surface of the casing; and the illumination device is arranged on both sides of the casing. The processor is an FPGA, and there are two illumination devices 1 (light sources). The hyperspectral camera specifically includes: an imaging lens 2 and an imaging unit (not shown in the figure). The imaging unit is separately connected to the imaging lens and the processor. There are two imaging lenses. The RGB DOF camera specifically includes: an infrared camera, a laser 6, and an RGB camera ; the infrared camera, the laser, and the RGB camera are all connected to a controller; and the laser is configured to obtain distances of different points in the image. There are two infrared cameras: an infrared camera A and an infrared camera B. A reference sign of the infrared camera A in FIG. 1 and FIG. 2 is 4, and a reference sign of the infrared camera B in FIG. 1 and FIG. 2 is 3. Specifically, the image target recognition system based on an RGB DOF camera and a hyperspectral camera provided by the present disclosure includes: an RGB camera, configured to obtain a two-dimensional image; a hyperspectral camera, configured to scan spectral information in line information of a target in the two-dimensional image; a processing unit, configured to match the spectral information obtained by the hyperspectral camera; and a PC, electrically connected to the processing unit and configured to display an image target; and further includes: a DOF unit, configured to trigger collection of a 3D DOF image based on the matched spectral information. The DOF unit includes an infrared camera A, an infrared camera B, and a laser. The laser is configured to obtain distances of different points in the image, and the infrared camera A, the infrared camera B, the laser, and the RGB camera are all configured to collect the 3D DOF image. The infrared camera A, the infrared camera B, the laser, and the RGB camera work together to collect the 3D DOF image. The DOF unit is self-adaptive, and can adjust a corresponding operating parameter based on the illuminance, to achieve accurate imaging under strong light or dark environments, provide better 3D DOF detail images, and further improve clarity of an image target. The system further includes a casing and light sources symmetrically arranged at both ends of the casing through a horizontal plate. The hyperspectral camera includes an imaging lens and an imaging unit fixedly connected to the imaging lens. The imaging lens is embedded at a symmetric center of the two light sources on the casing. The imaging unit is arranged inside the casing. Above the imaging lens, the infrared camera A, the laser, the RGB camera, and the infrared camera B are arranged in sequence from left to right. The processing unit is arranged inside the casing. The hyperspectral camera, the RGB camera, the infrared camera A, the infrared camera B, and the laser are all electrically connected to the processing unit to reduce data processing processes. The light sources are provided, and high-clarity DOF images are obtained through the DOF unit, avoiding the shortcoming of untimely triggering in harsh environments. Embodiment 2 All camera lenses are 25 mm prime lens. A processing unit is an FPGA. Transmission can be wired or wireless. The wired transmission is implemented through a USB signal line. The wireless transmission is implemented by performing corresponding matching through a wireless transmission unit of the FPGA. A DOF camera is triggered based on spectral information, which can reduce much data collection, analysis and processing work. Because a hyperspectral camera always performs line scan on a target to capture spectral information of the determined target, when a spectrum can be matched, a signal with a wavelength of 860 nm is used as a trigger signal, and the DOF camera is triggered quickly. A value of the signal at 860 nm is fed back to an infrared camera A4 and an infrared camera B3 at both ends, so that the infrared camera A4, the infrared camera B3, a laser 6, and an RGB camera 5 work together to obtain 3D image information of the determined target. Such responses are made at us level, and a DOF image of the recognized target is quickly obtained, which greatly improves the recognition efficiency, accuracy, and real-time performance.
As shown in FIG. 3, a highlighted target and its texture and other information can be clearly seen. Embodiment 3 A light source 1 is a light source with a range of 350 nm to 1000 nm. The light source 1 provides illumination, and allows a DOF camera to adjust an operating parameter based on illuminance, facilitating accurate imaging. An RGB camera 5 collects a two-dimensional image of a target; a hyperspectral camera scans line information of the target in the two-dimensional image to obtain spectral information of the target; a processing unit processes the spectral information of the target, and transmits the spectral information to a PC through a USB signal line; the PC displays the spectral information and matches it with a spectrum database to complete preliminary target recognition; spectral information with a wavelength of 860 nm is selected as a trigger condition from the spectral information on which the preliminary target recognition has been completed; an infrared camera A4, a laser 6, the RGB camera 5, and an infrared camera B3 in the DOF camera are triggered based on the spectral information with a wavelength of 860 nm to collect a 3D DOF image; and the 3D DOF image is processed by the processing unit and sent to the PC through the USB signal line for display, to complete final target recognition. The present disclosure solves a prior-art problem that when image target recognition is performed only through an RGB camera and a hyperspectral camera, an image target has low definition and is not highlighted, and image texture information cannot be obtained. The present disclosure can quickly and accurately recognize a target and provide a high-clarity recognized image. In this specification, several specific examples are used for illustration of the principles and implementations of the present disclosure. The description of the foregoing embodiments is used to help illustrate the method of the present disclosure and the core ideas thereof. In addition, those of ordinary skill in the art can make various modifications in terms of specific implementations and scope of application in accordance with the ideas of the present disclosure. In conclusion, the content of this specification should not be construed as a limitation to the present disclosure.