A lightless interior cavity machine vision system and imaging method

Abstract

The application discloses a light-free internal cavity machine vision system, an imaging method and an anaerobic or hypoxic workstation. The light-free internal cavity is constructed by a full-black solid PMMA plate and is used for providing a closed experimental space for blocking visible light. An infrared camera array includes multiple light-free infrared cameras which are installed on the inner wall of the light-free internal cavity. The imaging optical axes of the light-free infrared cameras converge at the center point of the bottom plate of the light-free internal cavity and are used for collecting multi-view infrared images inside the light-free internal cavity in a light-free environment. An image processing unit is connected with the infrared camera array and is used for fusing the multi-view infrared images to generate a simulated stereovision field. A head-mounted display device is connected with the image processing unit and is internally provided with an orientation detection device which is used for acquiring the visual orientation information of a wearer in real time and outputting the visual orientation information to the image processing unit. The application realizes eye-hand coordination experimental operation.

Description

Technical Field

[0001] This invention relates to the field of biological experimental equipment technology, and in particular to a lightless internal cavity machine vision system, imaging method, and an anaerobic or hypoxic workstation containing the system. Background Technology

[0002] An anaerobic or hypoxic workstation is an airtight chamber device used to simulate anaerobic or hypoxic environments. It stabilizes the experimental chamber in an anaerobic or hypoxic state through controlled chemical reactions or gas replacement, and is widely used in life science fields such as anaerobic microbial culture and somatic cell research. To facilitate observation and operation by researchers, existing anaerobic or hypoxic workstations are typically designed with a highly transparent PMMA observation window and a white PMMA inner chamber that facilitates diffuse reflection.

[0003] In the life sciences, the effects of visible light (wavelength range of 380nm to 780nm) on metabolism have received widespread attention. Visible light can drive the generation of free radicals, which can significantly impact metabolism by disrupting process molecules, leading to alterations in metabolic pathways, disease, and even death. The adverse effects of visible light on life have been defined as phototoxicity. However, while existing anaerobic or hypoxic workstations meet the needs of researchers for visual observation, their structural design exposes experimental subjects within the chamber to continuous visible light irradiation, resulting in serious phototoxicity problems.

[0004] On the one hand, transparent PMMA has a visible light transmittance of over 92%, allowing a large amount of light, whether from natural light or laboratory lighting, to penetrate the inner cavity. This exposes anaerobic microorganisms and somatic cells to a high dose of visible light, generating a large number of free radicals during their metabolism. The strong oxidizing power of these free radicals directly alters the chemical composition of many metabolic molecules, preventing normal metabolic processes and resulting in severe phototoxicity. On the other hand, white PMMA has a transmittance of only about 4%, allowing a considerable amount of visible light to penetrate the inner cavity from sides other than the observation panel. This further increases the visible light dose within the cavity, amplifying the adverse effects of phototoxicity on the experimental subjects. Furthermore, the high transmittance of white PMMA significantly reduces the effectiveness of light-blocking treatments on the transparent observation panel; even completely blocking visible light from the observation panel cannot eliminate the adverse effects of phototoxicity on the experimental subjects.

[0005] To reduce the effects of visible light, existing measures include using black fabric covers to shield the entire chamber from light during operation, or wrapping sample containers with aluminum foil. However, these measures only provide temporary light protection during non-operational periods. Once the personnel need to operate the equipment and remove the light shield or aluminum foil, visible light re-illuminates the sample, and the irreversible phototoxic effects generated during operation cannot be eliminated by re-shielding. Furthermore, repeated shielding and unshielding increases operational complexity and the risk of internal contamination.

[0006] In summary, existing anaerobic or hypoxic workstations rely on visible light reflection to form visual perception, which cannot simultaneously meet the contradiction between the need for a completely dark experimental environment and the need for visual observation and operation by experimental personnel, thus limiting the accuracy and reliability of anaerobic microorganism and somatic cell research. Summary of the Invention

[0007] The purpose of this invention is to provide a light-free internal cavity machine vision system, which solves the problems of existing technology, provides real-time stereo vision, and fundamentally resolves the contradiction between visible light phototoxicity and experimental personnel's observation and operation.

[0008] To achieve the above technical objectives, this invention provides a light-free cavity machine vision system, comprising: a light-free cavity constructed of solid black PMMA sheet to provide a closed experimental space that blocks visible light; an infrared camera array comprising multiple light-free infrared cameras mounted on the inner wall of the light-free cavity, wherein the imaging optical axes of each light-free infrared camera converge at the center point of the bottom plate of the light-free cavity, for acquiring multi-view infrared images of the interior of the light-free cavity in a light-free environment; an image processing unit connected to the infrared camera array for fusing the multi-view infrared images to generate a simulated stereoscopic visual field; and a head-mounted display device connected to the image processing unit, having a built-in orientation detection device for acquiring the wearer's visual orientation information in real time and outputting it to the image processing unit, wherein the image processing unit generates a simulated visual image with a corresponding viewpoint from the simulated stereoscopic visual field based on the visual orientation information and outputs it to the head-mounted display device for display.

[0009] The present invention provides the above-mentioned technical solution. The present invention utilizes the extremely low visible light transmittance of solid black PMMA sheet to construct a completely light-free experimental cavity, fundamentally eliminating phototoxicity. At the same time, a dome-shaped infrared camera array is used to collect multi-view infrared images in a light-free environment. The images are then fused by artificial intelligence to generate a simulated stereoscopic visual field. The operator's visual orientation is then acquired in real time through a head-mounted display device, and the corresponding simulated visual image is output. This allows the operator to obtain a real-time stereoscopic visual experience equivalent to that of the naked eye in a completely light-free cavity, realizing eye-hand coordinated experimental operation.

[0010] As a further improvement, the infrared camera array is arranged in a dome shape, including a top camera array, a front camera array, and a rear camera array; the top camera array is mounted on the inner surface of the top plate of the light-free cavity, the front camera array is mounted on the upper half of the inner surface of the front panel of the light-free cavity, and the rear camera array is mounted on the upper half of the inner surface of the back plate of the light-free cavity.

[0011] As a further improvement, the top camera array consists of 48 light-free infrared cameras arranged in a 16×3 array, the front camera array consists of 32 light-free infrared cameras arranged in a 16×2 array, and the rear camera array consists of 32 light-free infrared cameras arranged in a 16×2 array, for a total of 112 light-free infrared cameras.

[0012] As a further improvement, the image processing unit runs artificial intelligence stereoscopic field of view processing software, which fuses the multi-view infrared images to generate the simulated stereoscopic visual field, and renders the simulated visual image of the corresponding viewpoint in real time according to the visual orientation information fed back by the head-mounted display device.

[0013] As a further improvement, the AI ​​stereoscopic field-of-view processing software has a video image analysis mode. In this mode, the software actively learns about the operator's hand, objects, and equipment within the light-free cavity and automatically establishes a field-of-view image recognition model.

[0014] As a further improvement, each of the light-free infrared cameras is mounted on the inner wall of the light-free cavity via a universal adjustment bracket; the universal adjustment bracket includes a camera bracket base, a horizontal angle axis and a vertical angle axis, the camera bracket base is fixed to the inner wall of the light-free cavity, and the horizontal angle axis and the vertical angle axis are used to adjust the imaging optical axis direction of the light-free infrared camera.

[0015] As a further improvement, the universal adjustment bracket also includes a camera mounting adapter, which secures the light-free infrared camera with screws and connects to the horizontal and vertical angle axes via camera mounting screws; the horizontal and vertical angle axes are locked in place via bidirectional angle axis fixing screws after angle adjustment is completed.

[0016] As a further improvement, the head-mounted display device is a VR glasses, and the orientation detection device includes a three-dimensional gyroscope and a microgravity sensing system, used to acquire the visual orientation and direction information of the wearer's head in real time.

[0017] As a further improvement, the all-black solid PMMA sheet has a transmittance of less than 0.05% for visible light in the wavelength range of 380nm to 780nm.

[0018] As a further improvement, the fields of view of each non-light infrared camera in the infrared camera array overlap, forming an effective visual field with multiple imaging angles.

[0019] Correspondingly, the present invention also provides an anaerobic or hypoxic workstation, which adopts a light-free cavity machine vision system provided by the present invention. The light-free cavity is the airtight experimental cavity of the anaerobic or hypoxic workstation. The front panel of the light-free cavity is provided with an operation hole, which is used to install a sealing sleeve so that the operator can operate from the outside.

[0020] Accordingly, the present invention also provides a machine vision imaging method for a light-free cavity, applied to a light-free cavity constructed from solid black PMMA sheet, comprising the following steps: acquiring multi-view infrared images of the interior of the light-free cavity in a light-free environment using an infrared camera array mounted on the inner wall of the light-free cavity, wherein the imaging optical axes of each infrared camera converge at the center point of the bottom plate of the light-free cavity; fusing the multi-view infrared images to generate a simulated stereoscopic visual field using an image processing unit; acquiring the wearer's visual orientation information in real time using a head-mounted display device; and generating a simulated visual image with a corresponding viewpoint from the simulated stereoscopic visual field based on the visual orientation information, and outputting it to the head-mounted display device for display.

[0021] As a further improvement, the image processing unit runs artificial intelligence stereoscopic field-of-view processing software, which performs fusion processing on the multi-view infrared images and has a video image discrimination mode; in the video image discrimination mode, it actively learns the operator's hand, objects and equipment in the light-free cavity, automatically establishes a field-of-view image recognition model and incorporates it into the daily operation process.

[0022] As a further improvement, the head-mounted display device is a VR headset, with a built-in 3D gyroscope and microgravity sensing system; the 3D gyroscope and the microgravity sensing system acquire the wearer's visual orientation and direction information in real time and digitally output it to the image processing unit.

[0023] This invention constructs a completely light-free experimental chamber using solid black PMMA sheets, fundamentally eliminating the phototoxic effects of visible light on anaerobic microorganisms and somatic cells. This allows the cultivation and detection of experimental subjects to be carried out in a pristine, light-free environment, ensuring the authenticity and reliability of experimental data. By employing a dome-shaped infrared camera array combined with artificial intelligence image fusion, a stereoscopic visual field covering the entire operating area is constructed in the light-free environment, solving the fundamental problem of researchers being unable to observe operations under light-free conditions. The VR glasses' built-in 3D gyroscope and microgravity sensing system track the operator's gaze in real time, simulating the natural changes in the visual image as the head moves, providing the operator with an experience equivalent to naked-eye observation. Through AI video image analysis, the system can actively learn and update its model for different experimental scenarios, demonstrating strong sustainable application capabilities. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the infrared array vision system structure in a light-free cavity according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the working principle of stereoscopic vision imaging in an embodiment of the present invention; Figure 3 This is a schematic diagram of the dome-shaped camera array and the effective simulated visual field in an embodiment of the present invention; Figure 4 This is a structural diagram of the cavity of the dome camera array in an embodiment of the present invention; Figure 5 This is a schematic diagram of the imaging optical axis of the dome camera array in an embodiment of the present invention; Figure 6 This is a schematic diagram of the appearance of the light-free infrared camera in an embodiment of the present invention; Figure 7 This is a schematic diagram of the universal adjustment bracket in an embodiment of the present invention; Figure 8 This is a schematic diagram of the installation of the light-free infrared camera in an embodiment of the present invention; Figure 9 This is a schematic diagram illustrating the installation of VR glasses and an anaerobic or hypoxic workstation in an embodiment of the present invention.

[0025] Figure 10 This is the lightless internal cavity machine vision operation process of the anaerobic or hypoxic workstation in the embodiments of the present invention.

[0026] Reference numerals: 1. Dark infrared camera; 2. Front panel; 3. Imaging optical axis; 4. Operating hole; 5. Top plate; 6. Dark cavity; 7. Back plate; 8. Dark cavity bottom plate; 9. Simulated stereoscopic vision field; 10. Operator's hand; 11. Sealing sleeve; 12. Dome-shaped infrared camera array; 121. Top camera array; 122. Front camera array; 123. Rear camera array; 13. Left side panel; 14. Right side panel; 15. Center point; 16. Camera base; 17. Camera lens; 18. Camera mounting screw; 19. Camera base fastening screw; 20. Horizontal corner axis; 21. Bidirectional corner axis fixing screw; 22. Camera bracket base; 23. Vertical corner axis; 24. Camera mounting adapter; 25. Workstation host; 26. HDMI interface; 27. HDMI dual-connection data cable; 28. VR glasses. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] like Figures 1 to 10 As shown, this invention provides a light-free cavity machine vision system, comprising: a light-free cavity 6, constructed from solid black PMMA sheet, for providing a closed experimental space that blocks visible light; an infrared camera array, including multiple light-free infrared cameras 1, mounted on the inner wall of the light-free cavity 6, with the imaging optical axes 3 of each camera converging at the center point 15 of the bottom plate 8 of the light-free cavity, for acquiring multi-view infrared images of the interior of the light-free cavity in a light-free environment; an image processing unit, connected to the infrared camera array, for fusing the multi-view infrared images to generate a simulated stereoscopic visual field; and a head-mounted display device, connected to the image processing unit, with a built-in orientation detection device for acquiring the wearer's visual orientation information in real time and outputting it to the image processing unit. The image processing unit generates a simulated visual image with a corresponding viewpoint from the simulated stereoscopic visual field based on the visual orientation information and outputs it to the head-mounted display device for display.

[0029] The present invention provides the above-mentioned technical solution. The present invention utilizes the extremely low visible light transmittance of solid black PMMA board to construct a completely light-free experimental cavity, fundamentally eliminating phototoxicity. At the same time, a dome-shaped infrared camera array 12 is used to collect multi-view infrared images in a light-free environment. The images are then fused by artificial intelligence to generate a simulated stereoscopic visual field. The operator's visual orientation is then acquired in real time through a head-mounted display device, and the corresponding simulated visual image is output. This allows the operator to obtain a real-time stereoscopic visual experience equivalent to that of the naked eye in a completely light-free cavity, realizing eye-hand coordinated experimental operation.

[0030] As a further improvement, the infrared camera array is arranged in a dome shape, including a top camera array, a front camera array, and a back camera array; the top camera array is installed on the inner surface of the top plate 5 of the light-free cavity, the front camera array is installed on the upper half of the inner surface of the front panel 2 of the light-free cavity, and the back camera array is installed on the upper half of the inner surface of the back plate 7 of the light-free cavity.

[0031] As a further improvement, the top camera array consists of 48 light-free infrared cameras arranged in a 16×3 array, the front camera array consists of 32 light-free infrared cameras arranged in a 16×2 array, and the rear camera array consists of 32 light-free infrared cameras arranged in a 16×2 array, for a total of 112 light-free infrared cameras.

[0032] As a further improvement, the image processing unit runs artificial intelligence stereoscopic field of view processing software, which fuses the multi-view infrared images to generate the simulated stereoscopic visual field, and renders the simulated visual image of the corresponding viewpoint in real time according to the visual orientation information fed back by the head-mounted display device.

[0033] As a further improvement, the AI ​​stereoscopic field-of-view processing software has a video image analysis mode. In this mode, the software actively learns about the operator's hand, objects, and equipment within the light-free cavity and automatically establishes a field-of-view image recognition model.

[0034] As a further improvement, each of the light-free infrared cameras is mounted on the inner wall of the light-free cavity via a universal adjustment bracket; the universal adjustment bracket includes a camera bracket base, a horizontal angle axis and a vertical angle axis, the camera bracket base is fixed to the inner wall of the light-free cavity, and the horizontal angle axis and the vertical angle axis are used to adjust the imaging optical axis direction of the light-free infrared camera.

[0035] As a further improvement, the universal adjustment bracket also includes a camera mounting adapter, which secures the light-free infrared camera with screws and connects to the horizontal and vertical angle axes via camera mounting screws; the horizontal and vertical angle axes are locked in place via bidirectional angle axis fixing screws after angle adjustment is completed.

[0036] As a further improvement, the head-mounted display device is a VR glasses, and the orientation detection device includes a three-dimensional gyroscope and a microgravity sensing system, used to acquire the visual orientation and direction information of the wearer's head in real time.

[0037] As a further improvement, the all-black solid PMMA sheet has a transmittance of less than 0.05% for visible light in the wavelength range of 380nm to 780nm.

[0038] As a further improvement, the fields of view of each non-light infrared camera in the infrared camera array overlap, forming an effective visual field with multiple imaging angles.

[0039] Correspondingly, the present invention also provides an anaerobic or hypoxic workstation, which adopts a light-free cavity machine vision system provided by the present invention. The light-free cavity is the airtight experimental cavity of the anaerobic or hypoxic workstation. The front panel 2 of the light-free cavity is provided with an operation hole 4. The operation hole 4 is used to install a sealing sleeve 11 so that the operator can operate from the outside.

[0040] Accordingly, the present invention also provides a machine vision imaging method for a light-free cavity, applied to a light-free cavity constructed from solid black PMMA sheet, comprising the following steps: acquiring multi-view infrared images of the interior of the light-free cavity in a light-free environment using an infrared camera array mounted on the inner wall of the light-free cavity, wherein the imaging optical axes of each infrared camera converge at the center point 15 of the bottom plate 8 of the light-free cavity; fusing the multi-view infrared images to generate a simulated stereoscopic visual field using an image processing unit; acquiring the wearer's visual orientation information in real time using a head-mounted display device; and generating a simulated visual image with a corresponding viewpoint from the simulated stereoscopic visual field based on the visual orientation information, and outputting it to the head-mounted display device for display.

[0041] As a further improvement, the image processing unit runs artificial intelligence stereoscopic field-of-view processing software, which performs fusion processing on the multi-view infrared images and has a video image discrimination mode; in the video image discrimination mode, it actively learns the operator's hand, objects and equipment in the light-free cavity, automatically establishes a field-of-view image recognition model and incorporates it into the daily operation process.

[0042] As a further improvement, the head-mounted display device is a VR glasses with a built-in three-dimensional gyroscope and a microgravity sensing system; the three-dimensional gyroscope and the microgravity sensing system acquire the visual orientation and direction information of the wearer's head in real time and output it digitally to the image processing unit.

[0043] This invention constructs a completely light-free experimental chamber using solid black PMMA sheets, fundamentally eliminating the phototoxic effects of visible light on anaerobic microorganisms and somatic cells. This allows the cultivation and detection of experimental subjects to be carried out in a pristine, light-free environment, ensuring the authenticity and reliability of experimental data. A dome-shaped infrared camera array 12, combined with artificial intelligence image fusion, constructs a stereoscopic visual field covering the entire operating area in the light-free environment, solving the fundamental problem of researchers being unable to observe operations under light-free conditions. The VR glasses' built-in 3D gyroscope and microgravity sensing system track the operator's gaze in real time, simulating the natural changes in the visual image as the head moves, providing the operator with an experience equivalent to naked-eye observation. Through AI video image analysis, the system can actively learn and update its model for different experimental scenarios, demonstrating strong sustainable application capabilities.

[0044] This invention addresses the deficiency of effective visual capabilities in a dark environment in ordinary anaerobic or hypoxic workstations. It employs an infrared dark imaging camera array with overlapping camera video fields of view, creating a blind-spot-free visual information source within the dark cavity. High-performance computing and AI algorithms synthesize the visual information from the camera array into real-time, blind-spot-free, full-range intracavity visual information. The infrared dark camera array collects visual information from 180° left and right and 160° up and down angles within the anaerobic or hypoxic workstation cavity. Using VR glasses, the AI ​​algorithms utilize the 3D gyroscope and microgravity sensor system within the VR glasses to acquire the wearer's visual direction and angle in real time. Based on the data from the human eye's visual range, a simulated scene is displayed in real-time within the VR glasses' field of view, providing the experimenter with a visual experience equivalent to that of the naked eye, thus enabling experimental observation and operation within the dark cavity.

[0045] To address the lack of effective light-blocking measures in ordinary anaerobic or hypoxic workstations, this design utilizes solid black PMMA material to create a completely dark experimental environment that integrates experimental operations and cultivation. Inside the light-blocked workstation, an infrared dark-light camera array is recessed onto the observation panel with operating holes and its left and right side panels, following the operator's preferred observation method. The imaging axes of all the infrared dark-light cameras converge at the center point 15 of the workstation's inner cavity floor, forming a spherical visual dome structure centered on this center point. The workstation uses a Windows-based computer as its control center, sharing the same computer for both control and visual operation, thus avoiding the complex space requirements of multiple computers. Finally, the workstation includes a data interface for VR glasses, facilitating VR glasses connectivity and use.

[0046] In such Figures 1 to 10 In the preferred embodiment shown, the present invention: I. Construction of the Lightless Cavity The all-black solid PMMA workstation features a dark interior constructed from solid black PMMA sheets, including a top panel 5, front panel 2, left side panel, right side panel, back panel 7, and a dark interior bottom panel 8. Solid black PMMA is a plastic material that absorbs and blocks visible light across the entire 380nm to 780nm wavelength range, with a visible light transmittance of less than 0.05%. The dark interior made of solid black PMMA can simulate the native dark environment of anaerobic microorganisms and somatic cells, fundamentally eliminating the adverse effects of phototoxicity on experimental subjects. However, since 380nm to 780nm is the visible light range perceptible to the human eye, the solid black PMMA material blocks light in this wavelength range, making it impossible for researchers to directly observe the interior with the naked eye. Therefore, an infrared camera array and machine vision system are required to achieve visual observation in the dark environment.

[0047] The inner cavity is airtight, and the front panel has an operating hole for installing a sealing sleeve. The operator can perform experimental operations while maintaining the airtightness of the inner cavity through the sealing sleeve.

[0048] II. Infrared Camera Array like Figure 1 , Figure 3 , Figure 4 and Figure 5As shown, the infrared camera array is arranged in a dome shape and installed on the inner wall of the light-free cavity, forming a dome-shaped infrared camera array 12. Specifically, the dome-shaped infrared camera array 12 includes a top camera array 121, a front camera array 122, and a rear camera array 123. The top camera array is installed on the inner surface of the top plate of the light-free cavity and consists of 48 light-free infrared cameras arranged in a 16×3 (16 columns × 3 rows) array; the front camera array is installed on the upper half of the inner surface of the front panel of the light-free cavity and consists of 32 light-free infrared cameras arranged in a 16×2 (16 columns × 2 rows) array; the rear camera array is installed on the upper half of the inner surface of the back plate of the light-free cavity and consists of 32 light-free infrared cameras arranged in a 16×2 array. The camera array contains a total of 112 light-free infrared cameras.

[0049] like Figure 5 As shown, the imaging optical axes of all the light-free infrared cameras converge at the center point 15 of the light-free inner cavity base plate 8. This design creates a spherical visual dome structure centered on the center point of the inner cavity base plate. The fields of view of each light-free infrared camera overlap, as shown... Figure 3 As shown, an effective visual field with multiple imaging angles is formed. In the absence of light, the objects in the cavity and the operator's hand are imaged on the image sensor of the infrared camera due to the different amounts of infrared radiation. Since the camera array has multiple overlapping fields of view, all the field of view information can be aggregated to form a stereoscopic visual information network covering the entire operating area.

[0050] III. Camera Mounting Bracket like Figures 6 to 8 As shown, each light-free infrared camera is mounted on the inner wall of the light-free cavity via a universal adjustment bracket. It includes: camera base 16, camera lens 17, camera mounting screw 18, camera base fastening screw 19, horizontal angle shaft 20, bidirectional angle shaft fixing screw 21, camera bracket base 22, vertical angle shaft 23, and camera mounting adapter 24.

[0051] The universal adjustable bracket includes a camera bracket base, a horizontal angle axis, a vertical angle axis, a camera mounting adapter, and related fixing screws. The camera bracket base is fixed to a designated mounting position on the inner wall of the light-free cavity using screws. The light-free infrared camera is fixed to the camera mounting adapter using screws, and the camera mounting adapter is connected to the horizontal and vertical angle axes using camera fixing screws. During installation, the laser positioner is fitted onto the lens of each light-free infrared camera. By adjusting the horizontal and vertical angle axes, the laser spot is positioned at the center point of the base plate to ensure precise convergence of the imaging optical axis. Then, the fixing screws of the bidirectional angle axes are tightened, and the camera base fastening screws are tightened simultaneously to ensure that the light-free infrared camera is accurately and securely mounted on the inner wall.

[0052] All data cables of the camera array are connected via a dedicated TYPE-C interface, and all cables are routed along the internal cable trays to the control computer of the anaerobic or hypoxic workstation.

[0053] IV. Image Processing Unit The image processing unit is a high-performance computer equipped with artificial intelligence stereo field processing software. This computer also serves as the main control computer for the anaerobic or hypoxic workstation, uniformly controlling the gas environment control and machine vision system of the workstation, eliminating the complex space occupation of multiple computers.

[0054] The AI-powered stereoscopic field-of-view processing software fuses multi-view infrared images captured by all non-light infrared cameras, synthesizing overlapping multi-angle visual information into a simulated stereoscopic visual field. When the operator wears VR glasses, the software uses the VR glasses' built-in 3D gyroscope and microgravity sensor system to acquire the operator's head orientation and direction information in real time. Based on this orientation and direction information, it renders simulated visual images from the corresponding viewpoints in real time from the simulated stereoscopic visual field and outputs them to the VR glasses' display screen.

[0055] The AI-powered stereoscopic field-of-view processing software features a video image recognition mode. In this mode, the software can perform differential recognition and active learning on operator hands, objects within the cavity, and experimental equipment in a dark environment, automatically building a field-of-view image recognition model and incorporating it into daily operations. This function enables the system to adapt to different experimental scenarios and operating habits, providing excellent sustainable application capabilities.

[0056] V. Head-mounted display devices The head-mounted display device is VR glasses 28. For example... Figure 9 As shown, the VR glasses 28 are connected to the HDMI interface 26 on the control panel of the anaerobic or low-oxygen workstation host 25 via an HDMI dual-connection data cable 27. The VR glasses have a built-in 3D gyroscope and microgravity sensing system, which can acquire the wearer's head's visual orientation and direction information in real time, and digitally output the relevant positioning information to the image processing unit. Based on this information, the image processing unit renders a simulated stereoscopic image of the inner cavity scene with the corresponding viewpoint and perspective from the simulated stereoscopic visual field in real time, and outputs it to the VR glasses display. The operator wearing the VR glasses can obtain real-time stereoscopic vision in a light-free inner cavity, and the visual image changes naturally with the operator's head movements, achieving an operating experience equivalent to naked-eye observation.

[0057] VI. System Assembly and Operation Procedures The assembly process of the light-free cavity machine vision system is as follows: First, construct the light-free cavity of the anaerobic or low-oxygen workstation using solid black PMMA sheets, and assemble it airtight according to the process drawings. Then, fix the camera bracket base to the corresponding mounting position on the inner wall of the light-free cavity, and install the light-free infrared cameras on each camera bracket. Use laser locators to fit each camera lens one by one, adjust the bracket's corner axis until the laser spot falls on the center point of the base plate, and tighten the corner axis fixing screws. Connect the dedicated data cable to the interface of each camera, and run the cable along the internal cable tray to the main control computer. Install the artificial intelligence stereoscopic field of view processing software on the computer, and input and fuse the video signals of all the light-free infrared cameras. Connect the VR glasses to the main control computer via HDMI data cable, so that the software can acquire the orientation and direction signal data of the VR glasses in real time. At the same time, the software outputs the simulated cavity stereoscopic scene image to the VR glasses display based on the orientation and direction signal data. After completing the electrical connection, test the communication connectivity. The artificial intelligence learns and trains on the experimental operation scenario, and it is ready for normal use.

[0058] VII. Work Process Analysis Combination Figures 1 to 10 The working process of this invention is as follows: like Figure 4 , Figure 5 As shown, the anaerobic or low-oxygen workstation enclosure is constructed entirely of solid black PMMA sheets, forming a light-free interior cavity. An infrared camera array is mounted on the inner surfaces of the front panel 2, top panel 5, and back panel 7.

[0059] like Figure 5 As shown, the imaging optical axes of 112 infrared cameras without light converge at the center point of the base plate, forming a dome-shaped camera array. The fields of view of each camera overlap, and all field of view information is aggregated in the image processing unit and fused by AI algorithms to form a stereoscopic simulated visual field.

[0060] like Figure 3 As shown, the dome-shaped camera array forms an effective simulated visual field covering the inner cavity operating area. The operator's hand is inserted into the inner cavity through a sealed sleeve, and the dome-shaped camera array acquires infrared images of the operating scene from multiple angles.

[0061] The operator wears VR glasses, such as Figure 9 As shown, the VR glasses' built-in 3D gyroscope and microgravity sensor system provide real-time feedback on the wearer's visual position and orientation, and digitally output the relevant positioning information. Based on the positioning and orientation data, the AI ​​software simulates a three-dimensional image of the inner cavity scene with the corresponding viewpoint and perspective, and outputs it to the VR glasses' display, enabling the VR glasses wearer to obtain real-time stereoscopic vision in the dark inner cavity.

[0062] Figure 10This is the lightless internal cavity machine vision operation process of the anaerobic or hypoxic workstation in the embodiments of the present invention.

[0063] To meet the need for establishing real-time tracking vision in the dark cavity of an anaerobic or hypoxic workstation, this invention employs a dark infrared camera to construct an array machine vision system within the dark cavity of an anaerobic or hypoxic workstation.

[0064] Technical Solution: An infrared light field array imaging system consisting of 112 lightless infrared cameras is constructed inside the anaerobic or hypoxic workstation. The imaging optical axes of each lightless infrared camera converge at the center point of the workstation's base plate. Through the overlapping of the camera array's fields of view, an effective visual field with multiple imaging angles is formed. The visual field information is fused using AI artificial intelligence software algorithms. Based on the wearer's visual orientation and direction obtained from the VR glasses' 3D gyroscope and microgravity sensor information, a simulated visual image at that visual orientation and direction is displayed on the VR glasses' screen. This simulated visual image is updated in real time as the VR glasses wearer's visual orientation and direction change. Ultimately, this achieves real-time tracking machine vision for the operator of the anaerobic or hypoxic workstation, helping the operator to observe the interior of the lightless anaerobic or hypoxic workstation in real time and perform eye-hand coordinated operations.

[0065] Working Principle: The anaerobic or low-oxygen workstation enclosure is constructed entirely of solid black PMMA sheets on all six sides, forming a light-free interior cavity. An array of 112 light-free infrared cameras forms a vision system on the front, top, and back panels of the cavity. The optical axes of each camera converge at the center point of the workstation's base, forming a dome-shaped camera array. This creates a surround-angle stereoscopic field of view with overlapping perspectives. Due to differences in infrared radiation, the interior cavity and its contents are imaged on the infrared camera sensors in the light-free environment. Because of the overlapping perspectives of the camera array, all visual information is aggregated by AI software and fused into a three-dimensional, effective simulated visual field using AI algorithms. The operator wears VR glasses. The built-in 3D gyroscope and microgravity device provide real-time feedback on the wearer's visual position and orientation, digitizing the relevant positioning information. Based on the VR glasses' positioning and orientation data, the AI ​​software simulates the interior scene from the corresponding viewpoint and perspective, outputting a stereoscopic image to the VR glasses' display, providing the wearer with real-time stereoscopic vision within the light-free cavity.

[0066] Structural features: a. The interior of the anaerobic or hypoxic workstation is constructed entirely of solid black PMMA sheets. The matte interior includes a top panel, front panel, left side panel, right side panel, and bottom panel. The front panel is equipped with an operating hole.

[0067] b. A top camera array is distributed on the inner surface of the light-free inner cavity top plate, consisting of 48 light-free infrared cameras arranged in a 16×3 array, with the imaging optical axes of the 48 cameras converging at the center point of the bottom plate.

[0068] c. The upper half of the inner surface of the front panel of the light-free cavity is equipped with a front camera array, consisting of 32 light-free infrared cameras arranged in a 16×2 array, with the imaging optical axes of the 32 cameras converging at the center point of the base plate.

[0069] d. The upper half of the inner surface of the back plate of the lightless cavity is covered with a back camera array, consisting of 32 lightless infrared cameras arranged in a 16×2 array, with the imaging optical axes of the 32 cameras converging at the center point of the base plate.

[0070] e. Secure the camera bracket base to the designated mounting position on the inner wall of the light-free cavity using screws. Secure the infrared camera to the camera mounting adapter using screws. Then, screw the camera mounting screws on the bracket into the adapter's screw holes, and simultaneously tighten the camera base securing screws to ensure the light-free infrared camera is accurately and securely mounted on the solid black PMMA board on the inner wall of the light-free cavity. Place the laser positioner in front of the lens and adjust the horizontal and vertical corner axes on the bracket to position the laser point at the center point of the base plate. Tighten the bidirectional corner axis fixing screws, and then insert the TYPE-C dedicated data cable into the camera interface. Connect all data cables of the camera array to the control computer of the anaerobic or hypoxic workstation.

[0071] f. Insert one end of the HDMI dual-connection cable into the VR glasses interface and the other end into the HDMI interface on the anaerobic or low-oxygen workstation control panel.

[0072] In a preferred embodiment of the present invention, a light-free interior cavity is constructed for the anaerobic or hypoxic workstation. The light-free interior cavity of the anaerobic or hypoxic workstation is constructed using solid black PMMA sheets. The front panel, top panel, left side panel, right side panel, and bottom panel of the anaerobic or hypoxic workstation are made of solid black PMMA sheets and assembled into a complete, airtight enclosure.

[0073] In a preferred embodiment of the present invention, an artificial intelligence-based light-free intracavity machine vision system is constructed: a. Install 112 camera brackets into the corresponding mounting positions on the inner wall of the light-free cavity, and fix the bracket holders to the inner wall of the light-free cavity with screws; b. Secure 112 non-light infrared cameras to each camera bracket with screws; c. Place the laser positioner onto the lens of each infrared camera without light, adjust the bracket's angle axis until the laser spot falls on the center point of the base plate, and tighten the angle axis fixing screws; d. Connect 112 dedicated TYPE-C data cables to the TYPE-C adapter on each camera, and route the cables along the internal cable trays to the main control computer of the anaerobic or hypoxic workstation. e. Install AI-powered stereoscopic field-of-view processing software on the computer to fuse video signals from all infrared cameras without light. f. Connect the HDMI dual-connection data cable to the VR glasses and the HDMI interface of the anaerobic or low-oxygen workstation host respectively, so that the VR glasses can connect with the AI ​​artificial intelligence stereoscopic field processing software, so that the software can obtain the orientation and direction signal data of the VR glasses in real time. At the same time, the software outputs the simulated internal cavity stereoscopic field image to the VR glasses display based on the orientation and direction signal data. g. The artificial intelligence control software has a video image recognition mode. In this mode, it can actively learn from the differences between the operator's hands, internal objects, and internal experimental equipment, and automatically build a field-of-view image recognition model, which is then incorporated into the daily operation process.

[0074] In a preferred embodiment of the present invention, the assembly of a light-free internal cavity machine vision system in an anaerobic or hypoxic workstation is as follows: Assemble the anaerobic or hypoxic workstations according to the assembly process, perform airtight installation and electrical connections as required by the process drawings, connect the AI ​​control software to the main control system, and test the communication connectivity. The AI ​​learns and trains on the experimental operation scenarios.

[0075] Compared with the prior art, the present invention has the following significant advantages: In terms of materials, this invention uses solid black PMMA to construct a light-free inner cavity, with a visible light transmittance of less than 0.05%. In contrast, existing technologies use transparent PMMA with a transmittance of over 92% and white PMMA with a transmittance of approximately 4%, neither of which can provide effective light protection for the experimental subject. Regarding the vision system, this invention employs a light-free infrared camera array to form machine vision. Through AI software and VR glasses, it achieves real-time stereoscopic scene display within the effective imaging area, perfectly matching the state of human vision. Existing technologies rely on visible light reflection to form human vision, failing to provide any visual capability in a dark environment. In terms of control, this invention uses an artificial intelligence control system that synchronously controls the system in real time through VR glasses. The operator only needs to wear VR glasses to complete automatic tracking and observation, indistinguishable from direct human observation. Existing technologies lack corresponding control methods.

[0076] This invention provides a light-free cavity machine vision system and an anaerobic or hypoxic workstation containing the system. A completely light-blocking cavity is constructed using solid black PMMA sheet material. A dome-shaped camera array consisting of multiple light-free infrared cameras is installed on the inner wall of the cavity. The imaging optical axes of each camera converge at the center point of the cavity floor, forming a stereoscopic visual information acquisition network with overlapping fields of view. An image processing unit runs artificial intelligence stereoscopic field-of-view processing software to fuse multi-view infrared images and generate a simulated stereoscopic visual field. The operator wears VR glasses, whose built-in 3D gyroscope and microgravity sensing system acquire the operator's visual orientation and direction information in real time. Based on this information, the image processing unit renders simulated visual images from the corresponding viewpoints in real time from the simulated stereoscopic visual field and outputs them to the VR glasses for display. This allows the operator to obtain real-time stereoscopic vision equivalent to that of the naked eye within the completely light-free cavity, enabling eye-hand coordinated experimental observation and operation.

[0077] This invention employs a light-free machine vision system in the field of anaerobic or hypoxic workstations, increasing the ability of such devices to provide a completely light-free environment for experiments. A light-free infrared camera array drives the acquisition of the light-free infrared field of view within the anaerobic or hypoxic workstation cavity. The light-free infrared camera bracket is specially designed for installation within the anaerobic or hypoxic workstation cavity. VR glasses are also designed for use in the light-free cavity operation within the anaerobic or hypoxic workstation field.

[0078] This invention's light-free internal cavity machine vision system enables anaerobic or hypoxic workstations to achieve a completely light-free experimental environment. This allows for the cultivation and experimental detection of anaerobic microorganisms and somatic cells to achieve the original environmental state, thereby enabling the successful cultivation of light-sensitive biological samples and facilitating the study of the actual metabolic state of experimental subjects. It fundamentally breaks through the technical bottleneck in anaerobic microorganism and somatic cell science research, providing reliable technical support for cutting-edge research in life sciences.

[0079] The machine vision system with a light-free cavity of this invention enables anaerobic or hypoxic workstations to achieve a completely light-free experimental environment, providing the possibility for in-depth research on photosensitive proteins. It fundamentally overcomes the problem of serious deviation in experimental data caused by the inability to effectively shield visible light, and provides major technical support for cutting-edge brain-computer interface research, light-controlled blindness treatment research, and targeted anti-cancer research. The lightless intracavity machine vision system of this invention adopts an AI artificial intelligence algorithm architecture, which can learn, train and upgrade for special experimental scenarios, avoiding interruption of use caused by the complexity of scientific research experiments, and laying a good foundation for its sustainable application.

[0080] It should be understood that the scope of protection sought by this invention is not limited to the non-limiting embodiments, which are merely illustrative examples. The substantive scope of protection claimed in this application is further embodied in the scope provided by the independent claims and their dependent claims.

Claims

1. A lightless intracavity machine vision system, characterized in that, include: The light-free interior cavity is constructed from solid black PMMA sheets to provide a closed experimental space that blocks visible light. An infrared camera array, comprising multiple light-free infrared cameras, is installed on the inner wall of the light-free cavity. The imaging optical axes of each light-free infrared camera converge at the center point of the bottom plate of the light-free cavity, and are used to acquire multi-view infrared images of the interior of the light-free cavity in a light-free environment. An image processing unit, connected to the infrared camera array, is used to fuse the multi-view infrared images to generate a simulated stereoscopic visual field. A head-mounted display device is connected to the image processing unit and has a built-in orientation detection device for acquiring the wearer's visual orientation information in real time and outputting it to the image processing unit. The image processing unit generates a simulated visual image with a corresponding viewpoint from the simulated stereoscopic visual field based on the visual orientation information and outputs it to the head-mounted display device for display.

2. The light-free intracavity machine vision system according to claim 1, characterized in that: The infrared camera array is arranged in a dome shape, including a top camera array, a front camera array, and a rear camera array; the top camera array is installed on the inner surface of the top plate of the light-free cavity, the front camera array is installed on the upper half of the inner surface of the front panel of the light-free cavity, and the rear camera array is installed on the upper half of the inner surface of the back plate of the light-free cavity.

3. The light-free intracavity machine vision system according to claim 2, characterized in that: The top camera array consists of 48 infrared cameras arranged in a 16×3 array, the front camera array consists of 32 infrared cameras arranged in a 16×2 array, and the rear camera array consists of 32 infrared cameras arranged in a 16×2 array, for a total of 112 infrared cameras.

4. The light-free intracavity machine vision system according to claim 1, characterized in that: The image processing unit runs AI stereoscopic field-of-view processing software, which fuses the multi-view infrared images to generate the simulated stereoscopic visual field, and renders the simulated visual image of the corresponding viewpoint in real time based on the visual orientation information fed back by the head-mounted display device.

5. The light-free intracavity machine vision system according to claim 4, characterized in that: The AI-powered stereoscopic field-of-view processing software has a video image analysis mode. In this mode, the software actively learns about the operator's hands, objects, and equipment within the light-free cavity and automatically builds a field-of-view image recognition model.

6. The light-free intracavity machine vision system according to claim 1, characterized in that: Each of the light-free infrared cameras is mounted on the inner wall of the light-free cavity via a universal adjustment bracket; the universal adjustment bracket includes a camera bracket base, a horizontal angle axis and a vertical angle axis, the camera bracket base is fixed to the inner wall of the light-free cavity, and the horizontal angle axis and the vertical angle axis are used to adjust the imaging optical axis direction of the light-free infrared camera.

7. The light-free intracavity machine vision system according to claim 6, characterized in that: The universal adjustment bracket also includes a camera mounting adapter, which secures the light-free infrared camera with screws and connects to the horizontal and vertical angle axes via camera mounting screws; the horizontal and vertical angle axes are locked in place via bidirectional angle axis fixing screws after angle adjustment is completed.

8. The light-free intracavity machine vision system according to claim 1, characterized in that: The head-mounted display device is a VR glasses, and the orientation detection device includes a three-dimensional gyroscope and a microgravity sensing system, which are used to acquire the visual orientation and direction information of the wearer's head in real time.

9. The light-free intracavity machine vision system according to claim 1, characterized in that: The solid black PMMA sheet has a transmittance of less than 0.05% for visible light in the wavelength range of 380nm to 780nm.

10. A light-free intracavity machine vision system according to claim 1, characterized in that: The fields of view of each non-light infrared camera in the infrared camera array overlap, forming an effective visual field with multiple imaging angles.

11. An anaerobic or hypoxic workstation, characterized in that: The system includes a light-free cavity machine vision system according to any one of claims 1 to 10, wherein the light-free cavity is the airtight experimental cavity of the anaerobic or hypoxic workstation, and the front panel of the light-free cavity is provided with an operation hole for installing a sealing sleeve so that the operator can operate from the outside.

12. A light-free cavity machine vision imaging method, applied to a light-free cavity constructed from solid black PMMA sheet, characterized in that, Includes the following steps: An array of infrared cameras installed on the inner wall of the dark cavity acquires multi-view infrared images of the interior of the dark cavity in a dark environment, wherein the imaging optical axes of each infrared camera converge at the center point of the bottom plate of the dark cavity. The multi-view infrared images are fused by the image processing unit to generate a simulated stereoscopic visual field. The wearer's visual orientation information is obtained in real time through a head-mounted display device; The image processing unit generates a simulated visual image from the simulated stereoscopic visual field based on the visual orientation information, and outputs it to the head-mounted display device for display.

13. The light-free intracavity machine vision imaging method according to claim 12, characterized in that: The image processing unit runs AI stereoscopic field-of-view processing software, which performs fusion processing on the multi-view infrared images and has a video image discrimination mode. In the video image recognition mode, the operator's hands, objects and equipment in the light-free cavity are actively learned, and a field-of-view image recognition model is automatically established and incorporated into the daily operation process.

14. The light-free intracavity machine vision imaging method according to claim 12, characterized in that: The head-mounted display device is a VR glasses, with a built-in three-dimensional gyroscope and a microgravity sensing system; the three-dimensional gyroscope and the microgravity sensing system acquire the visual orientation and direction information of the wearer's head in real time and output it digitally to the image processing unit.

Images