Depth camera and imaging system

Abstract

The utility model aims to provide a depth camera and an imaging system. The depth camera comprises at least three camera devices which at least comprise a first camera device, a second camera device and a third camera device, and any two of the first camera device, the second camera device and the third camera device form a binocular vision camera. The at least three camera devices at least provide a first binocular vision camera and a second binocular vision camera, a connecting line between optical centers of two camera devices forming the binocular vision camera is defined as a base line of the binocular vision camera, the first binocular vision camera is provided with a first base line, the second binocular vision camera is provided with a second base line, and the first base line is provided with a second base line. The first base line and the second base line are arranged in a non-collinear manner; and the processor inputs the picture signals collected by the camera device into the processor, and the picture signals are processed by the processor to obtain a fused depth picture.

Description

Technical Field

[0001] This application relates to depth cameras and imaging systems. Background Technology

[0002] With the continuous exploration of computer vision and the ongoing development of visual algorithms, depth cameras have emerged. Also known as 3D cameras, depth cameras are cameras capable of acquiring distance information of objects in a scene. They are widely used in 3D reconstruction, object detection, and environmental perception. Unlike traditional 2D cameras, depth cameras can acquire depth information by capturing the spatial environment.

[0003] A binocular depth camera is a device based on the principle of binocular stereo vision. It uses two cameras to capture the same scene from different angles and calculates the depth information of objects using parallax. This technology is similar to human eyes, perceiving the distance of objects by comparing the differences between the images seen by the left and right eyes. Furthermore, binocular depth cameras generally do not actively emit light, hence they are also called passive depth cameras.

[0004] However, stereo depth cameras cannot detect weak texture stripes parallel to the camera baseline. In real life, weak texture stripes are quite common, such as solid-colored table edges and fences. When faced with such objects, stereo depth cameras cannot correctly determine the depth value and therefore output invalid depth regions.

[0005] In indoor office areas, plain-colored desks, lockers, and chairs are everywhere. Due to the limitations of binocular depth cameras, many invalid depth areas appear when outputting depth information for these areas. Therefore, solving the problem that binocular depth cameras cannot output depth information for weakly textured strip-shaped objects is of great significance to the development of indoor computer vision.

[0006] Therefore, there is a need in the art for a new depth camera and imaging system to solve one or a combination of the above problems. Utility Model Content

[0007] The purpose of this application is to overcome the problem that binocular depth cameras cannot output depth information for weakly textured strip-shaped objects, based on the structural principle of using binocular vision to realize a depth camera.

[0008] A depth camera according to a first aspect of this application includes: at least three camera devices, including at least a first camera device, a second camera device, and a third camera device, wherein any two of the first camera device, the second camera device, and the third camera device constitute a binocular vision camera, the at least three camera devices provide at least a first binocular vision camera and a second binocular vision camera, a line connecting the optical centers of two camera devices constituting the binocular vision camera is defined as the baseline of the binocular vision camera, the first binocular vision camera has a first baseline, the second binocular vision camera has a second baseline, and the first baseline and the second baseline are non-collinearly arranged; a processor, which processes the data... The image signals acquired by at least three camera devices are input to the processor, and the processor processes them to obtain a fused depth image. A driving device is connected to at least one of the first, second, and third camera devices to drive the camera devices connected to the driving device to move, so that the first baseline and / or the second baseline move. Optionally, the first binocular vision camera acquires a first binocular depth image signal, and the second binocular vision camera acquires a second binocular depth image signal. The acquired first and second binocular depth image signals are input to the processor, and the processor processes them to obtain a fused depth image.

[0009] The beneficial effects of the above embodiments include, but are not limited to, that, based on the structural principle of realizing a depth camera through binocular vision, by adding at least one camera device to the depth camera to form at least two binocular vision cameras, and the baselines of the two binocular vision cameras are not collinear, the problem that the binocular depth camera cannot output depth information for weak texture strip-shaped objects is solved, and the structure with movable baselines can change the effective depth range of the depth camera through the design of baseline movement.

[0010] In one or more embodiments of the depth camera, the first baseline is perpendicular to the second baseline.

[0011] The beneficial effects of the above embodiments include, but are not limited to, the inventors' discovery that the structure in which the first baseline is perpendicular to the second baseline can better solve the problem that weak texture strip-shaped objects of binocular depth cameras cannot output depth information compared to other angles.

[0012] In one or more embodiments of the depth camera, the first camera device, the second camera device, and the third camera device are all connected to the driving device. The first camera device and the second camera device constitute a first binocular vision camera and move in a first direction, while the third camera device moves in a second direction.

[0013] In one or more embodiments of the depth camera, the first direction is perpendicular to the second direction.

[0014] In one or more embodiments of the depth camera, a mounting platform is further included. A transmission mechanism is provided on the mounting platform. The first camera device, the second camera device, and the third camera device are mounted on the mounting platform via the transmission mechanism. The first camera device and the second camera device are mounted on the first transmission mechanism extending along a first direction, and the third camera device is mounted on the second transmission mechanism extending along a second direction. Optionally, the length of the first transmission mechanism is 100mm-800mm, the length of the second transmission mechanism is 100mm-800mm, and the movement of the first baseline and the second baseline allows the length of the baseline to be adjustable within a range of 10mm-400mm. Optionally, the first transmission mechanism and the second transmission mechanism are selected from a slide rail, an adjusting screw, a lead screw, or a guide rail slide. Optionally, the first transmission mechanism is a slide rail, and the second transmission mechanism is an adjusting screw.

[0015] In one or more embodiments of the depth camera, the first camera device is fixed in place, and the second and third camera devices are both connected to the driving device. The first and second camera devices constitute a first binocular vision camera, and the first and third camera devices constitute a second binocular vision camera. The second camera device moves in a first direction, and the third camera device moves in a second direction. Optionally, the first direction is perpendicular to the second direction.

[0016] In one or more embodiments of the depth camera, a mounting platform is further included. A transmission mechanism is provided on the mounting platform. The second and third camera devices are mounted on the mounting platform via the transmission mechanism. The first camera device is fixed on the mounting platform. The second camera device is mounted on a third transmission mechanism extending along a first direction. The third camera device is mounted on a fourth transmission mechanism extending along a second direction. Optionally, the lengths of the third and fourth transmission mechanisms are 100mm-800mm, and the movement of the first and second baselines allows the length of the baseline to be adjustable within a range of 10mm-400mm. Optionally, the third and fourth transmission mechanisms are selected from slide rails, adjusting screws, lead screws, or guide rail slides. Optionally, the third transmission mechanism is a first slide rail, and the fourth transmission mechanism is a second slide rail.

[0017] In one or more embodiments of the depth camera, a fixing mechanism is further included, the fixing mechanism including a clamp, wherein the first camera device, the second camera device, and the third camera device are disposed on the mounting platform via the clamp, and the clamp is mounted and connected to the mounting platform;

[0018] The first camera device, the second camera device, and the third camera device include RGB cameras or monochrome cameras with the same focal length parameters.

[0019] In one or more embodiments of the depth camera, the imaging device includes: a beam splitter capable of splitting a light beam input to the beam splitter into multiple light beams with different output directions; multiple cameras corresponding to receiving the light beams in each output direction and outputting electrical signals, each camera including: a lens corresponding to receiving the light beams in each output direction; a photosensitive element corresponding to receiving the light beams output by the lens and converting them into electrical signals, the output direction of the electrical signals output by the photosensitive element being configured to point to a processor; for the multiple cameras, at least a first camera corresponding to a first depth of field and a second camera corresponding to a second depth of field are included, the values ​​of the first depth of field and the second depth of field being different.

[0020] An imaging system according to a second aspect of this application has a depth camera as described in the first aspect, the imaging system including an indoor monitoring system and a stereo vision industrial imaging system. Attached Figure Description

[0021] The above and other features, properties and advantages of this application will become more apparent from the following description taken in conjunction with the accompanying drawings and embodiments, wherein:

[0022] Figure 1 This is a schematic diagram of the structure of a depth camera according to one embodiment.

[0023] Figure 2 This is a schematic diagram of the structure of a depth camera according to another embodiment.

[0024] Figure 3 This is a schematic diagram of the imaging device of a depth camera according to an embodiment.

[0025] Figure label:

[0026] 100-Depth Camera

[0027] 1-Camera device, 11-First camera device, 12-Second camera device, 13-Third camera device, 101-First binocular vision camera, 1001-First baseline, 102-Second binocular vision camera, 1002-Second baseline, 111-Lens, 112-Photosensitive element; 110-First camera, 120-Second camera;

[0028] 2-Processor;

[0029] 3-Drive device;

[0030] 4- Mounting platform, 41- First transmission mechanism, 42- Second transmission mechanism, 43- Third transmission mechanism, 44- Fourth transmission mechanism;

[0031] 51- Beam splitter;

[0032] 2011 - First optical axis, 2012 - Second optical axis, 2001 - First beam, 2002 - Second beam. Detailed Implementation

[0033] The following discloses various embodiments or examples of the subject matter technical solutions. To simplify the disclosure, specific examples of elements and arrangements are described below. Of course, these are merely examples and are not intended to limit the scope of protection of the present invention. For example, a first feature subsequently described in the specification being formed above or on a second feature may include an embodiment in which the first and second features are formed in a direct connection, or an embodiment in which an additional feature is formed between the first and second features, so that the first and second features are not directly connected. In addition, reference numerals and / or letters may be repeated in different examples in these disclosures. This repetition is for brevity and clarity and does not in itself indicate a relationship between the various embodiments and / or structures to be discussed. Furthermore, when the first element is described in a manner connected or combined with the second element, the description includes embodiments in which the first and second elements are directly connected or combined with each other, as well as embodiments in which one or more other intervening elements are added to indirectly connect or combine the first and second elements with each other.

[0034] It is understood that the following flowcharts are used to illustrate the steps performed by the cultivation method according to embodiments of this application. It should be understood that, depending on the actual situation, the preceding or following steps may not necessarily be performed precisely in order. Other steps may be added to these processes, or one or more steps may be removed from them.

[0035] It is understood that the terms "depth camera" and "binocular vision camera" in this case refer to a concept encompassing multiple camera devices. The first, second, and third camera devices can be a single camera or a combination of multiple cameras. However, a single camera device outputs a 2D image without depth, while a non-depth camera can output a 3D image with depth. Although those skilled in the art may use the terms "camera" and "camera" interchangeably, in this application, "depth camera" and "binocular vision camera" refer to a concept encompassing multiple camera devices, not a single camera device.

[0036] Those skilled in the art will understand that each camera device may include one camera or multiple cameras. A single camera generally includes a lens and a photosensitive element. The lens is used to receive a light beam. The photosensitive element is used to receive the light beam output by the lens and convert it into an electrical signal. The output direction of the electrical signal output by the photosensitive element is configured to point to the processor 2.

[0037] refer to Figure 1 as well as Figure 2As shown, the depth camera 100 includes at least three camera devices 1, a processor 2, and a drive device 3.

[0038] like Figure 1 as well as Figure 2 As shown, at least three camera devices 1 include at least a first camera device 11, a second camera device 12, and a third camera device 13. Any two of the first camera device 11, the second camera device 12, and the third camera device 13 constitute a binocular vision camera. The at least three camera devices provide at least a first binocular vision camera 101 and a second binocular vision camera 102. The line connecting the optical centers of the two camera devices constituting the binocular vision camera is defined as the baseline of the binocular vision camera. The first binocular vision camera 101 has a first baseline 1001, and the second binocular vision camera 102 has a second baseline 1002. The first baseline 1001 and the second baseline 1002 are non-collinearly arranged. Generally, the location of the optical center is the light transmission center of the camera device. For example, in computer vision, the optical center of a single camera is closely related to the camera's intrinsic parameter matrix. The intrinsic parameter matrix is ​​usually expressed as:

[0039]

[0040] Among them, c x and c y These are the coordinates of the principal optical center of the image, which is the projection position of the optical center onto the image sensor, and is usually close to the center of the image.

[0041] It can be understood that, theoretically, any two of the first camera device 11, the second camera device 12, and the third camera device 13 can form a binocular vision camera, providing three binocular vision cameras. The first camera device 11 and the second camera device 12 form one binocular vision camera, the second camera device 12 and the third camera device 13 form one binocular vision camera, and the first camera device 11 and the third camera device 13 form one binocular vision camera. However, in some scenarios, two sets can be selected to form two binocular vision cameras, which can reduce the computational requirements of the processor. But it can be understood that at least two binocular vision cameras need to be formed.

[0042] The signal flow between the processor 2 and the camera device 1 can be such that the image signals acquired by at least three camera devices 1 are input into the processor 2, and the fused depth image is obtained after processing by the processor 2.

[0043] Optionally, the signal flow between the processor 2 and the camera device 1 can specifically be as follows: the first binocular vision camera 101 acquires the first binocular depth image signal, the second binocular vision camera 102 acquires the second binocular depth image signal, and the acquired first binocular depth image signal and second binocular depth image signal are input into the processor 2, and the fused depth image is obtained after processing by the processor 2.

[0044] It can be understood that, for example, the first camera device 11, the second camera device 12, and the third camera device 13 can respectively acquire the first 2D image signal, the second 2D image signal, and the third 2D image signal, and input the first 2D image signal, the second 2D image signal, and the third 2D image signal into the processor 2, and obtain the fused depth image after processing by the processor 2.

[0045] The imaging device includes cameras that can be either RGB or monochrome. RGB cameras are used to acquire rich color information to enhance the visual details and texture recognition of the scene. Monochrome cameras can provide higher sensitivity and contrast in low-light conditions, ensuring accurate image capture in various lighting environments. Preferably, at least three imaging devices 1 are RGB or monochrome cameras configured with the same focal length parameters.

[0046] The method by which processor 2 fuses the first binocular depth image signal and the second binocular depth image signal is well known to those skilled in the art. For example, it can be to combine the RGB image corresponding to the depth map, register the depth map using the ORB method, obtain the homography matrix corresponding to the RGB image, and then perform preliminary fusion after registration.

[0047] Binocular depth image signals can be obtained using binocular depth measurement methods commonly used in this field. For example, the binocular vision camera is first calibrated, and the original image is corrected based on the calibration results. The two corrected images are located on the same plane and are parallel to each other. Pixel matching is performed on the two corrected images, and the depth of each pixel is calculated based on the matching results to obtain a depth map.

[0048] It is understood that any implementation method that processes the image data obtained from each camera device to obtain the depth information of the scene area can be applied to the solution provided in this application, and there are no restrictions on it.

[0049] It is understood that processor 2 is not limited to a specific processor. For example, in some cases, processor 2 can have a distributed structure, such as including processors located at the depth camera end and the backend cloud, with the image fusion processing described above being implemented by the depth camera end and / or the backend cloud. Furthermore, in embodiments employing a distributed structure, the specific execution terminal for each step can be adjusted according to actual conditions, and the specific implementation scheme of each step on a particular terminal should not limit the scope of protection of this application.

[0050] A drive device 3, connected to at least one of the first camera device 11, the second camera device 12, and the third camera device 13, drives the camera device connected to the drive device 3 to move, thereby moving the first baseline 1001 and / or the second baseline 1002. The movement is a change in the length and / or position of the baseline. For example... Figure 1 The driving device 3 is connected to the first camera device 11, the second camera device 12, and the third camera device 13, driving them to move, that is, driving the first baseline 1001 and the second baseline 1002 to move. The length of the first baseline 1001 changes, and the length and position of the second baseline 1002 also change. The first camera device 11 and the second camera device 12 constitute the first binocular vision camera 101, and either the first camera device 11 or the second camera device 12, together with the third camera device 13, constitute the second binocular vision camera 102. Or as... Figure 2 As shown, the driving device 3 is connected to the second camera device 12 and the third camera device 13, while the first camera device 11 is fixed. Driving both of them moves, that is, driving the first baseline 1001 and the second baseline 1002 to move, causing changes in the lengths of the first baseline 1001 and the second baseline 1002. In other words, the first camera device 11 and the second camera device 12 constitute the first binocular vision camera 101, and the first camera device 11 and the third camera device 13 constitute the second binocular vision camera 102. The driving device can be driven by common motors. For example, each of the first camera device 11, the second camera device 12, and the third camera device 13 can be equipped with a corresponding driving motor. The signal flow between the driving device 3 and the processor 2 can be an interaction between the motor controller and the processor 2. For example, the controller can control the motor to move the camera device 1 and record position information to obtain a complete target image. Alternatively, the processor 2 can directly interact with the motor; neither is a limitation.

[0051] The beneficial effects of the above embodiments include, but are not limited to: based on the structural principle of realizing a depth camera through binocular vision, by adding at least one camera device to the depth camera to form at least two binocular vision cameras, and the baselines of the two binocular vision cameras are not collinear, the problem that the binocular depth camera cannot output depth information for weak texture strip-shaped objects is solved, and the structure with movable baselines can change the effective depth range of the depth camera through the design of baseline movement.

[0052] refer to Figure 1 as well as Figure 2 As shown, in some embodiments, the first baseline 1001 is perpendicular to the second baseline 1002. The beneficial effect of this is that the inventors have found that the structure where the first baseline 1001 is perpendicular to the second baseline 1002 can better solve the problem of weak-textured strip-shaped objects in binocular depth cameras failing to output depth information compared to other angles. It can be understood that in... Figure 2In the illustrated embodiment, the first baseline 1001 and the second baseline 1002 remain perpendicular at all times, while... Figure 1 In the illustrated embodiment, the first baseline 1001 and the second baseline 1002 can be adjusted to a mutually perpendicular positional relationship within the range of baseline movement, or they can be adjusted to... Figure 1 The non-perpendicular positional relationship is shown in the figure.

[0053] refer to Figure 3 As shown, it can be understood that the camera device may include more than one camera. For example, in some embodiments, for camera device 1, it may include a beam splitter 51, which can split the light beam input to beam splitter 51 into multiple light beams with different output directions; multiple cameras, corresponding to receive the light beam in each output direction and output electrical signals. Each camera includes: a lens 111, which corresponds to receive the light beam in each output direction; a photosensitive element 112, which corresponds to receive the light beam output by lens 111 and convert it into an electrical signal. The output direction of the electrical signal output by photosensitive element 112 is configured to point to processor 2. For multiple cameras, at least a first camera 110 corresponding to a first depth of field and a second camera 120 corresponding to a second depth of field are included, and the values ​​of the first depth of field and the second depth of field are different.

[0054] The beam splitter 51 can have a rectangular parallelepiped structure, with the long side of the parallelepiped connecting the light-incident and light-exit surfaces of the beam splitter 51; this avoids obstructing the camera's field of view. Alternatively, the beam splitter 51 can have a cubic structure, and neither is a limitation.

[0055] The specific structure for beam splitting using the beam splitter 51 can be as follows: the beam splitter 51 has a beam splitting surface, and the beam splitting surface has a semi-transparent and semi-reflective coating; optionally, the semi-transparent and semi-reflective coating is a visible light coating with a beam splitting ratio of transmission:reflection = 50%:50%; optionally, the reflected light output by the beam splitter 51 corresponds to the first optical axis 2011, and the transmitted light output corresponds to the second optical axis 2012, wherein the first optical axis 2011 and the second optical axis 2012 are perpendicular, for example... Figure 3As shown, the first beam 2001 is reflected light, and its corresponding first optical axis 2011 is along the Y direction. The second beam 2002 is transmitted light, and its corresponding second optical axis 2012 is along the X direction. The X direction is perpendicular to the Y direction. This design allows for a clearer fused image. The semi-reflective beam splitter allows images from different cameras with varying depths of field to be fused by the image processor to obtain a large depth-of-field image. This overcomes the limitation of fixed-focus cameras in terms of limited depth of field, which restricts their applicability, while maintaining the simplicity and robustness of calibration. It is understood that the ratio of transmission to reflection can be adjusted according to the needs of the scene, not limited to the aforementioned 50%:50%. Furthermore, the beam splitting structure is not limited to having a semi-reflective coating on the splitting surface; other beam splitting structures are also possible.

[0056] For example Figure 3 As shown, the first camera 110 corresponds to the first depth of field, and the second camera 120 corresponds to the second depth of field. The values ​​of the first depth of field and the second depth of field are different. As introduced above, the meaning of depth of field here refers to the range of distances including the front and back of the subject that can be captured in a clear image at the front edge of the lens. That is, after focusing, the distance of the clear image presented within the range before and after the focal point is called the depth of field. Since the depth of field of the first camera 110 and the second camera 120 are different, the first image provided by the first camera 110 to the processor 2 and the second image output by the second camera 120 to the processor 2 are different. The processor 2 can fuse the first image and the second image to output a fused image, which is the image acquired by the camera device.

[0057] The specific fusion process can be as follows: For example, camera intrinsic parameter calibration can be performed using MATLAB software, and distortion correction can be achieved using distortion coefficients; simple image preprocessing can be performed, the perspective transformation matrix can be calculated, and the images from the two cameras can be registered; pixel-level image fusion can be performed on the registered images from the two cameras, and the final fused image can be output. The advantage of this approach is that by acquiring image data with different focus distances and field of view from various sensors and then performing fusion processing, it is possible to retain a larger range of images captured at a smaller focal length, while also fusing images from different focal lengths to obtain an image with a large depth of field and a wide field of view at the same moment.

[0058] refer to Figure 1 as well as Figure 2As shown, the depth camera 100 may also include a mounting platform 4. A transmission mechanism is installed on the mounting platform. The camera device can be directly fixed to the mounting platform, or it can be mounted on the mounting platform via the transmission mechanism. The transmission mechanism drives the camera device to move within a certain range, thereby changing the length and / or position of the baseline. The transmission mechanism can be a slide rail, adjusting screw, lead screw, or guide rail slide, but is not limited to these. The specific structure can be... Figure 1 As shown, the first camera device 11, the second camera device 12, and the third camera device 13 are mounted on the mounting platform 4 via the transmission mechanism. The first camera device 11 and the second camera device 12 are mounted on the first transmission mechanism 41 extending along a first direction, and the third camera device 13 is mounted on the second transmission mechanism 42 extending along a second direction. The length of the first transmission mechanism 41 is 100mm-800mm, and the length of the second transmission mechanism 42 is 100mm-800mm, for example, 120mm, 160mm, 200mm, 300mm, 400mm, 600mm, or 800mm. The movement of the first baseline 1001 and the second baseline 1002 allows the length of the baseline to be adjustable within a range of 10mm-400mm, for example, 10-400mm, 10-300mm, 20-300mm, 20-200mm, 10-200mm, 30-100mm, or 30-150mm. Figure 1 As shown, the first transmission mechanism 41 can be a slide rail, and the second transmission mechanism 42 can be an adjusting screw. Such an arrangement is compact and easy to adjust. For details, please refer to the description of Embodiment 1 below.

[0059] It could also be Figure 2 As shown, the second camera device 12 and the third camera device 13 are mounted on the mounting platform 4 via a transmission mechanism. The first camera device 11 is fixed on the mounting platform 4. The second camera device 12 is mounted on the third transmission mechanism 43 extending along the first direction, and the third camera device 13 is mounted on the fourth transmission mechanism 44 extending along the second direction. The lengths of the third transmission mechanism 43 and the fourth transmission mechanism 44 are 100mm-800mm, for example, 120mm, 160mm, 200mm, 300mm, 400mm, 600mm, or 800mm. The movement of the first baseline 1001 and the second baseline 1002 allows the length of the baseline to be adjustable within a range of 10mm-400mm, for example, 10-400mm, 10-300mm, 20-300mm, 20-200mm, 10-200mm, 30-100mm, or 30-150mm. Figure 2As shown, the third transmission mechanism 43 and the fourth transmission mechanism 44 can both be slide rail structures, that is, the third transmission mechanism 43 is the first slide rail and the fourth transmission mechanism 44 is the second slide rail. This makes the arrangement structure compact and easy to adjust. For details, please refer to the description of Embodiment 2 below.

[0060] As described above, the transmission mechanism can be a slide rail, adjusting screw, lead screw, or guide rail slide, etc., and is not limited to these. For the structure using a lead screw, please refer to the description of Example 3 below. For the structure using a guide rail slide, please refer to the description of Example 4 below.

[0061] It can be understood that, preferably, the first direction is perpendicular to the second direction, that is... Figure 1 , Figure 2 The first direction is the X direction, and the second direction is the Y direction.

[0062] It is understood that in some embodiments, the depth camera 100 further includes a fixing mechanism, which includes a clamp. The first camera device 11, the second camera device 12, and the third camera device 13 are mounted on the mounting platform 4 via the clamp. The clamp is mounted and connected to the mounting platform 4. The specific structure of the clamp can be referred to the descriptions of Embodiments 1 and 2 below.

[0063] This application also provides an imaging system having the depth camera 100 described in the above embodiments. The imaging system includes an indoor monitoring system and a stereo vision industrial imaging system. Specific examples illustrate different imaging systems below.

[0064] Example 1

[0065] A depth camera 100, such as Figure 1As shown, the system includes a first camera device 11, a second camera device 12, a third camera device 13, a processor 2, and a drive device 3. The first camera device 11, the second camera device 12, and the third camera device 13 are all RGB camera devices with a focal length of 4mm. The drive device includes a moving mechanism and a fixing mechanism. The first camera device 11 and the second camera device 12 are mounted on a slide rail extending in a first direction, serving as a first transmission mechanism 41. The third camera device 13 is mounted on an adjusting screw extending in a second direction, serving as a second transmission mechanism 42. The camera device 1 is fixed by a camera device clamp, which is fixed to a base by a cylindrical pin. The first camera device 11 and the second camera device 12 are fixed to the slide rail via the base, and the third camera device 13 is fixed to the adjusting screw via the base. The controller controls the motor to drive the base of the fixing mechanism to move on the slide rail and the adjusting screw, and records the position information to obtain a complete target image. The slide rail is 400mm long, and the baseline adjustment range between the first camera device 11 and the second camera device 12 on the slide rail is 30-150mm. The adjusting screw is 160mm long, and the baseline adjustment range between the third camera device 13 and the first camera device 11 and the second camera device 12 is 30-150mm.

[0066] The processor 2 determines the depth information of the first binocular vision camera 101 based on the image data received from the first camera device 11 and the second camera device 12; it determines the depth information of the second binocular vision camera 102 based on the image data received from the first camera device 11 and the third camera device 13; and it registers and fuses the depth information of the first binocular vision camera 101 and the depth information of the second binocular vision camera 102 to obtain a more accurate depth map with more complete depth information.

[0067] Example 2

[0068] A trinocular depth camera, such as Figure 2 As shown, it includes a first camera device 11, a second camera device 12, a third camera device 13, a processor 2, and a drive device 3. The camera device 1 is... Figure 3 The structure shown features a beam splitter 51, which is a 28×20×20mm cuboid prism, elongated along both the light-incident and light-out surfaces to ensure it does not obstruct the field of view of the camera device. The prism contains a standard 50:50 visible light coating in the center, splitting the incident light beam into a first beam and a second beam.

[0069] The first camera device 11, the second camera device 12, and the third camera device 13 all include RGB cameras. The first focal length of the first camera in a single camera device is configured to be 1.94mm, and the second focal length of the second camera is configured to be 4mm. The first camera is located at the first beam and the second camera is located at the second beam.

[0070] The second camera device 12 is mounted on a first slide rail extending along a first direction, serving as a third transmission mechanism 43. The third camera device 13 is mounted on a second slide rail extending along a second direction, serving as a fourth transmission mechanism 44. The first camera device 11 is fixed to the mounting platform 4 and located at the connection between the first and second slide rails. The second camera device 12 moves along the X-axis direction on the third transmission mechanism 43 via a fixing mechanism, and the third camera device 13 moves along the Y-axis direction on the fourth transmission mechanism 44 via a fixing mechanism. The lengths of the third and fourth transmission mechanisms 43 and 44 are 160 mm. The baseline between the second or third camera device 13 and the first camera device 11 is adjustable within a range of 30-150 mm. The fixing mechanism includes a camera clamp and a base. The camera clamp is fixed to the base by a cylindrical pin, and the beam splitter is fixed in a groove in the base. The controller controls the motor to drive the base of the fixing mechanism to move on the slide rails and records position information to obtain a complete target image.

[0071] The processor 2 determines the depth information of the first binocular vision camera 101 based on the image data received from the first camera device 11 and the second camera device 12; it determines the depth information of the second binocular vision camera 102 based on the image data received from the first camera device 11 and the third camera device 13; and it registers and fuses the depth information of the first binocular vision camera 101 and the depth information of the second binocular vision camera 102 to obtain a more accurate depth map with more complete depth information.

[0072] Example 3

[0073] A trinocular depth camera, compared to Embodiments 1 and 2, differs in the structure of its transmission mechanism. Embodiment 3 uses a lead screw as the transmission mechanism. Specifically, the second camera device 12 is mounted on the first lead screw extending along a first direction, and the third camera device is mounted on the second lead screw extending along a second direction. The first camera device is fixed to the mounting platform and located at the connection between the first and second lead screws. The controller controls the motor to drive the lead screw to rotate, thereby moving the base of the fixing mechanism in the first direction and recording position information to obtain a complete target image. This structure is easy to control and achieves full automation.

[0074] Example 4

[0075] A trinocular depth camera, compared to Embodiments 1, 2, and 3, differs in the structure of its transmission mechanism. Embodiment 4 uses a guide rail slide as its transmission mechanism. Specifically, a second camera device is mounted on a first guide rail slide extending along a first direction, and a third camera device is mounted on a second guide rail slide extending along a second direction. The first camera device is fixed to a mounting platform and located at the connection between the first and second guide rail slides. A controller controls a motor to drive the guide rail slide to move, thereby moving the base of the fixing mechanism in the first direction and recording position information to obtain a complete target image.

[0076] In summary, the depth camera and imaging system described in the above embodiments have beneficial effects including, but not limited to: based on the structural principle of realizing a depth camera through binocular vision, by adding at least one imaging device to the depth camera to form at least two binocular vision cameras, and with the baselines of the two binocular vision cameras being non-collinear, the problem of the binocular depth camera being unable to output depth information for weakly textured strip-shaped objects is solved. Furthermore, the movable baseline structure allows for changes in the effective depth range of the depth camera through baseline movement design. In particular, the structure in which the first baseline 1001 and the second baseline 1002 are perpendicular can better solve the problem of the binocular depth camera being unable to output depth information for weakly textured strip-shaped objects compared to other angles.

[0077] While this application discloses preferred embodiments as described above, it is not intended to limit the scope of this application. Any changes and modifications can be made by those skilled in the art without departing from the spirit and scope of this application. Therefore, any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of this application without departing from the content of the technical solution of this application shall fall within the protection scope defined by the claims of this application.

Claims

1. A depth camera (100), characterized in that, include: At least three camera devices (1) are provided, including at least a first camera device (11), a second camera device (12), and a third camera device (13). Any two of the first camera device (11), the second camera device (12), and the third camera device (13) constitute a binocular vision camera. The at least three camera devices provide at least a first binocular vision camera (101) and a second binocular vision camera (102). The line connecting the optical centers of the two camera devices constituting the binocular vision camera is defined as the baseline of the binocular vision camera. The first binocular vision camera (101) has a first baseline (1001), and the second binocular vision camera (102) has a second baseline (1002). The first baseline (1001) and the second baseline (1002) are non-collinearly arranged. The processor (2) inputs the image signals collected by the at least three camera devices (1) into the processor (2), and obtains a fused depth image after processing by the processor (2); A drive device (3) is connected to at least one of the first camera device (11), the second camera device (12), and the third camera device (13) to drive the camera device connected to the drive device (3) to move such that the first baseline (1001) and / or the second baseline (1002) move. Optionally, the first binocular vision camera (101) acquires a first binocular depth image signal, and the second binocular vision camera (102) acquires a second binocular depth image signal. The acquired first binocular depth image signal and second binocular depth image signal are input into the processor (2), and the fused depth image is obtained after processing by the processor (2).

2. The depth camera (100) as described in claim 1, characterized in that, The first baseline (1001) is perpendicular to the second baseline (1002).

3. The depth camera (100) as described in claim 1, characterized in that, The first camera device (11), the second camera device (12), and the third camera device (13) are all connected to the driving device (3). The first camera device (11) and the second camera device (12) constitute a first binocular vision camera (101) and move in a first direction, while the third camera device (13) moves in a second direction.

4. The depth camera (100) as described in claim 3, characterized in that, The first direction is perpendicular to the second direction.

5. The depth camera (100) as described in claim 3, characterized in that, It also includes an installation platform (4), on which a transmission mechanism is provided. The first camera device (11), the second camera device (12), and the third camera device (13) are mounted on the installation platform (4) through the transmission mechanism. The first camera device (11) and the second camera device (12) are mounted on the first transmission mechanism (41) extending in the first direction, and the third camera device (13) is mounted on the second transmission mechanism (42) extending in the second direction. Optionally, the length of the first transmission mechanism (41) is 100mm-800mm, the length of the second transmission mechanism (42) is 100mm-800mm, and the movement of the first baseline (1001) and the second baseline (1002) makes the length of the baseline adjustable in the range of 10mm-400mm. Optionally, the first transmission mechanism (41) and the second transmission mechanism (42) are selected from slide rail, adjusting screw, lead screw or guide rail slide; Optionally, the first transmission mechanism (41) is a slide rail, and the second transmission mechanism (42) is an adjusting screw.

6. The depth camera (100) as described in claim 1, characterized in that, The first camera device (11) is fixed in place, and the second camera device (12) and the third camera device (13) are both connected to the driving device (3). The first camera device (11) and the second camera device (12) constitute a first binocular vision camera (101), and the first camera device (11) and the third camera device (13) constitute a second binocular vision camera (102). The second camera device (12) moves in a first direction, and the third camera device (13) moves in a second direction. Optionally, the first direction is perpendicular to the second direction.

7. The depth camera (100) as described in claim 6, characterized in that, It also includes an installation platform (4), on which a transmission mechanism is provided. The second camera device (12) and the third camera device (13) are mounted on the installation platform (4) through the transmission mechanism. The first camera device (11) is mounted on the installation platform (4) and fixed. The second camera device (12) is mounted on the third transmission mechanism (43) extending along the first direction. The third camera device (13) is mounted on the fourth transmission mechanism (44) extending along the second direction. Optionally, the lengths of the third transmission mechanism (43) and the fourth transmission mechanism (44) are 100mm-800mm, and the movement of the first baseline (1001) and the second baseline (1002) makes the length of the baseline adjustable in the range of 10mm-400mm. Optionally, the third transmission mechanism (43) and the fourth transmission mechanism (44) are selected from slide rails, adjusting screws, lead screws or guide rail slides; Optionally, the third transmission mechanism (43) is the first slide rail, and the fourth transmission mechanism (44) is the second slide rail.

8. The depth camera (100) as described in claim 5 or 7, characterized in that, It also includes a fixing mechanism, which includes a clamp. The first camera device (11), the second camera device (12), and the third camera device (13) are mounted on the mounting platform (4) by the clamp, and the clamp is mounted and connected to the mounting platform (4). The first camera device (11), the second camera device (12), and the third camera device (13) include RGB cameras or monochrome cameras with the same focal length parameters.

9. The depth camera (100) as claimed in claim 1, characterized in that, The camera device (1) includes: The beam splitter (51) is capable of splitting the light beam input to the beam splitter (51) into multiple beams with different output directions; Multiple cameras, each corresponding to receive a beam of light from a specific output direction, and output an electrical signal. Each camera includes: Lens (111), lens (111) is used to receive the beam of light in each output direction; The photosensitive element (112) receives the light beam output from the lens (111) and converts it into an electrical signal. The output direction of the electrical signal output by the photosensitive element (112) is configured to point to the processor (2). The plurality of cameras include at least a first camera corresponding to a first depth of field and a second camera corresponding to a second depth of field, wherein the values ​​of the first depth of field and the second depth of field are different.

10. An imaging system, characterized in that, The imaging system includes a depth camera (100) as described in any one of claims 1-9, and includes an indoor monitoring system and a stereo vision industrial imaging system.

Images