Underwater robot and camera protection shell detection method thereof

Abstract

This specification discloses a detection method and detection device for an underwater robot camera protective shell, the protective shell is arc-shaped, and the detection method includes the following steps: collecting a first image of the protective shell; performing image enhancement processing on the first image to obtain a second image; obtaining the number of closed areas and / or defect coefficient of the second image; judging whether the number of closed areas exceeds a first threshold, and / or judging whether the defect coefficient exceeds a second threshold; if the number of closed areas exceeds the first threshold and / or the defect coefficient exceeds the second threshold, the protective shell does not meet the requirements; otherwise, the protective shell meets the requirements. The detection method and detection device for the protective shell can detect the cleanliness or integrity of the arc-shaped protective shell.

Description

Technical Field

[0001] The present invention relates to the field of underwater exploration technology, and in particular to a detection method and a detection device for a protective shell of an underwater robot camera. Background Art

[0002] The description in this section merely provides background information related to the disclosure of this specification and does not constitute prior art.

[0003] In the field of underwater robot exploration, the camera is a vital component of the underwater robot or other monitoring system. The main technical means of underwater exploration is to shoot images or video information through the camera.

[0004] However, due to the complex underwater environment, the cameras installed in underwater robots or other monitoring systems are easily stained. Therefore, a protective shell is generally provided outside the camera to prevent the camera itself from being damaged by physical or environmental factors.

[0005] Many protective cases are made of hemispherical transparent material, which allows the camera to capture 180-degree external image content. However, due to the complexity of the external environment, especially when underwater robots and other equipment work in the water for a long time, not only will there be sediment accumulation on the outside of the protective case, but there will also be other aquatic organisms attached to the outside of the protective case. Therefore, in order to ensure the normal operation of the camera, it is necessary to check the cleanliness or integrity of the protective case.

[0006] It should be noted that the above introduction to the technical background is only for the convenience of providing a clear and complete description of the technical solutions of this specification and for the convenience of understanding by those skilled in the art. It cannot be considered that the above technical solutions are well known to those skilled in the art simply because they are described in the background technology section of this specification. Summary of the invention

[0007] In view of the deficiencies in the prior art, the purpose of this specification is to provide a method and a device for detecting a protective shell of an underwater robot camera, which can detect the cleanliness or integrity of the arc-shaped protective shell.

[0008] To achieve the above object, the embodiment of this specification provides a method for detecting a protective shell of an underwater robot camera, wherein the protective shell is arc-shaped, and the detection method comprises the following steps:

[0009] capturing a first image of the protective shell;

[0010] Performing image enhancement processing on the first image to obtain a second image;

[0011] Acquiring the number of closed areas and / or defect coefficient of the second image;

[0012] Determining whether the number of closed areas exceeds a first threshold, and / or determining whether the defect coefficient exceeds a second threshold;

[0013] If the number of closed areas exceeds the first threshold and / or the defect coefficient exceeds the second threshold, the protective shell does not meet the requirements;

[0014] Otherwise, the protective case meets the requirements.

[0015] As a preferred embodiment, the protective shell is hemispherical, and the diameter of the protective shell is 20 cm to 200 cm.

[0016] As a preferred embodiment, the protective shell has a detection area, and the detection area includes a plurality of sub-areas;

[0017] In the step of acquiring the first image of the protective shell, the first image of each of the sub-areas is acquired simultaneously;

[0018] In the step of performing image enhancement processing on the first image, image enhancement processing is performed on the first images of the sub-regions simultaneously and in parallel.

[0019] As a preferred implementation, the detection area includes four rectangular sub-areas, and each of the sub-areas has an equal width and an equal length.

[0020] As a preferred implementation, the image enhancement processing includes correction, filtering, quantization, and white balance.

[0021] As a preferred implementation, the step of performing image enhancement processing on the first image to obtain the second image includes:

[0022] Processing the first image with a correction algorithm to obtain a corrected image;

[0023] Interpolation processing is performed on the gaps of the corrected image to obtain the second image.

[0024] As a preferred implementation, the step of obtaining the number of closed areas of the second image includes:

[0025] detecting edges of the second image;

[0026] The closedness of the edge is detected, and the number of the closed areas is obtained.

[0027] As a preferred implementation, the step of detecting the edge of the second image includes:

[0028] calculating a gradient of the second image;

[0029] Processing the second image using a template matrix;

[0030] Calculate the final gradient magnitude and gradient direction.

[0031] As a preferred implementation, the step of obtaining the defect coefficient of the second image includes:

[0032] Get the directional derivatives of all pixels in the bright field image ;

[0033] Get the directional derivatives of all pixels in the dark field imaging image ;

[0034] Calculate the defect coefficient of the second image .

[0035] The embodiment of the present specification also provides a detection device for a protective shell, wherein the protective shell is arc-shaped, and the detection device comprises:

[0036] A collection module, used for collecting a first image of the protective shell;

[0037] A processing module, used for performing image enhancement processing on the first image to obtain a second image;

[0038] An acquisition module, used for acquiring the number of closed areas and / or defect coefficient of the second image;

[0039] a judgment module, used to judge whether the number of closed areas exceeds a first threshold value, and / or whether the defect coefficient exceeds a second threshold value; if the number of closed areas exceeds the first threshold value and / or the defect coefficient exceeds the second threshold value, the protective shell does not meet the requirements; otherwise, the protective shell meets the requirements;

[0040] The processing module is electrically connected to the collection module and the acquisition module respectively, and the judgment module is electrically connected to the acquisition module.

[0041] An underwater robot, comprising a camera and a protective shell for protecting the camera, wherein the protective shell is arc-shaped, and the underwater robot further comprises a detection device for detecting contamination of the protective shell; wherein the detection device comprises:

[0042] A collection module, used for collecting a first image of the protective shell;

[0043] A processing module, used for performing image enhancement processing on the first image to obtain a second image;

[0044] An acquisition module, used for acquiring the number of closed areas and / or defect coefficient of the second image;

[0045] a judgment module, used to judge whether the number of closed areas exceeds a first threshold value, and / or whether the defect coefficient exceeds a second threshold value; if the number of closed areas exceeds the first threshold value and / or the defect coefficient exceeds the second threshold value, the protective shell does not meet the requirements; otherwise, the protective shell meets the requirements; the underwater robot also includes a cleaning device for cleaning the protective shell when the protective shell does not meet the requirements;

[0046] The processing module is electrically connected to the collection module and the acquisition module respectively, and the judgment module is electrically connected to the acquisition module.

[0047] Beneficial effects: The detection method and detection device of the protective shell provided in this embodiment, after collecting the first image of the protective shell, perform image enhancement processing on the first image to obtain a second image, and the second image is closer to the actual curved surface of the protective shell. Then obtain the number of closed areas and / or defect coefficient of the second image. Among them, the number of closed areas can reflect the amount of attachments on the protective shell. The more closed areas there are, the more attachments there are on the protective shell, and the less the protective shell meets the requirements and needs to be cleaned or replaced; the defect coefficient can reflect the amount of defects on the protective shell. The larger the defect coefficient, the more defects there are on the protective shell, and the less the protective shell meets the requirements and needs to be replaced or repaired. Determine whether the number of closed areas exceeds the first threshold value, and / or determine whether the defect coefficient exceeds the second threshold value. If the number of closed areas exceeds the first threshold value and / or the defect coefficient exceeds the second threshold value, the protective shell does not meet the requirements, otherwise the protective shell meets the requirements. Therefore, the detection method and detection device of the protective shell provided in this embodiment can detect the cleanliness or integrity of the arc-shaped protective shell, thereby ensuring the normal operation of the camera in the protective shell.

[0048] With reference to the following description and the accompanying drawings, the specific embodiments of the present specification are disclosed in detail, indicating the manner in which the principles of the present specification can be adopted. It should be understood that the embodiments of the present specification are not limited in scope.

[0049] Features described and / or illustrated with respect to one embodiment may be used in the same or similar manner in one or more other embodiments, combined with features in other embodiments, or substituted for features in other embodiments.

[0050] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, integers, steps or components, but does not exclude the presence or addition of one or more other features, integers, steps or components. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative labor.

[0052] Figure 1 This is a flowchart of the steps of a method for detecting a protective shell provided in this embodiment;

[0053] Figure 2 This is a schematic diagram of the structure of a protective shell provided in this embodiment;

[0054] Figure 3 This is a schematic diagram of the structure of a protective shell detection area provided in this embodiment;

[0055] Figure 4 is a flow chart of steps for performing image enhancement processing on the first image in this embodiment;

[0056] Figure 5 A line chart comparing the processing efficiency of the prior art and the present embodiment when processing different deformation levels;

[0057] Figure 6 A schematic diagram of a closure detection process provided in this embodiment;

[0058] Figure 7 A schematic diagram of the structure of a lighting assembly provided in this embodiment;

[0059] Figure 8 This is a schematic structural diagram of a protective shell detection device provided in this embodiment;

[0060] Fig. 9 This is a schematic structural diagram of a cleaning device for an underwater robot camera assembly provided in this embodiment;

[0061] Fig.10 This is a schematic diagram of the structure of an underwater robot provided in this embodiment;

[0062] Fig.11 for Figure 2 The front view of the underwater robot;

[0063] Fig.12 A front view of the underwater robot showing the direction of rotation of the protective cover;

[0064] Fig.13 An enlarged view of the outer surface of the protective cover.

[0065] Description of reference numerals:

[0066] 10. Protective shell; 60. Acquisition module; 2. Processing module; 3. Acquisition module; 40. Judgment module; 5. Control element; 6. Concave lens; 7. Convex lens; 8. Baffle;

[0067] C1, C2, C3, C4: sub-areas. DETAILED DESCRIPTION

[0068] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0069] It should be noted that when an element is referred to as being "disposed on" another element, it may be directly on the other element or there may be another element centered thereon. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be another element centered thereon at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only and do not represent the only implementation method.

[0070] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this specification belongs. The terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this specification. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0071] See also Figure 1 This embodiment provides a method for detecting a protective shell 10 of an underwater robot camera, wherein the protective shell 10 is arc-shaped. The contamination detection method comprises the following steps:

[0072] Step S10: capturing a first image of the protective shell 10;

[0073] Step S20: performing image enhancement processing on the first image to obtain a second image;

[0074] Step S30: obtaining the number of closed areas and / or defect coefficient of the second image;

[0075] Step S40: determining whether the number of closed areas exceeds a first threshold, and / or determining whether the defect coefficient exceeds a second threshold;

[0076] Step S50: If the number of closed areas exceeds the first threshold and / or the defect coefficient exceeds the second threshold, the protective shell 10 does not meet the requirements;

[0077] Step S60: Otherwise, the protective shell 10 meets the requirements.

[0078] Since the protective shell 10 is arc-shaped, the detection method of the protective shell 10 provided in the present embodiment performs image enhancement processing on the collected first image, so that the obtained second image can be closer to the actual curved surface condition.

[0079] In step S30, the number of closed areas of the second image can be obtained, or the defect coefficient of the second image can be obtained, or both the number of closed areas and the defect coefficient can be obtained. The number of closed areas can reflect the amount of attachments on the protective shell 10. The attachments may be attached mud, particles and other substances. The more closed areas there are, the more attachments there are on the protective shell 10, the worse the cleanliness of the protective shell 10 is, the less it meets the requirements, and it needs to be cleaned or replaced. The defect coefficient can reflect the amount of defects on the protective shell 10. The defects may be cracks caused by physical collisions, or cracks caused by defects in the protective shell 10 itself. The larger the defect coefficient, the more defects there are on the protective shell 10, and the less it meets the requirements of the protective shell 10, and it needs to be replaced or repaired.

[0080] The detection method of the protective shell 10 provided in this embodiment can detect the cleanliness or integrity of the arc-shaped protective shell 10, thereby ensuring that the camera in the protective shell 10 works normally.

[0081] In step S10, a camera or other acquisition module 60 may be used to acquire images of the protective shell 10. Figure 2 As shown, the protective shell 10 is hemispherical. The present embodiment does not limit the material and size of the protective shell 10, which can be made of transparent glass. Considering the performance and processing time of the modules performing each step, the mechanical performance, etc., the area of ​​the protective shell 10 should not be too large. In one embodiment, the diameter of the protective shell 10 can be set to 20 cm to 200 cm.

[0082] In this embodiment, the protective shell 10 has a detection area, and the first image is the image corresponding to the detection area. Since the protective shell 10 is arc-shaped, the angles of different positions of the first image are different. If the number of closed areas and / or defect coefficients are directly obtained based on the first image, there will be a gap between the actual arc surface situation and the obtained data, so step S20 is required. The detection area can also be defined as OIA (Object Image Area).

[0083] Specifically, in order to make the detection method faster, the detection area may include multiple sub-areas. In step S10, the first images of the sub-areas are collected simultaneously; in step S20, image enhancement processing is performed on the first images of the sub-areas simultaneously and in parallel.

[0084] In this embodiment, the sub-areas may be divided according to the distance and angle between the arc surface of the protective shell 10 and the collection module 60. Preferably, Figure 3 As shown, the detection area may include four rectangular sub-areas C1, C2, C3, and C4, at which time the algorithm has the best performance. Each sub-area has the same width and length.

[0085] In step S20, the image enhancement process may include correction, filtering, quantization, white balance, etc. Step S20 may specifically include: processing the first image with a correction algorithm to obtain a corrected image; interpolating the gaps in the corrected image to obtain a second image. Interpolation processing can ensure the signal continuity requirements of subsequent algorithms.

[0086] like Figure 4 As shown, it is a specific embodiment of step S20. In this embodiment, the input first image passes through the image cache S103 and the pixel interpolation S104, and the second image can be output. The address management S101 is responsible for recording the address information of the memory corresponding to the sub-area, the cache write S102 is responsible for the write operation, and the cache read S105 is responsible for transmitting the processed data back to the image cache S103. The image cache S103 includes 4 caches, which store the data of 4 sub-images respectively. The pixel interpolation S104 performs interpolation processing before the image is output. The correction sub-image processing module S106 can be used to perform image enhancement, which can make the obtained second image closer to the actual situation of the curved protective shell 10.

[0087] In this embodiment, the first image captured can be converted into a 12-bit, 16-bit, or 20-bit image to be suitable for a variety of image compression formats. The four sub-regions are processed in parallel to perform image enhancement processing on the data of the 12-bit, 16-bit, and 20-bit first image, respectively, and the correction speed is significantly improved, which is particularly effective in image deformation processing. Figure 5 As shown, the left figure is a line graph of the average number of pixels processed per unit time when the ordinary process processes different deformation levels, and the right figure is a line graph of the average number of pixels processed per unit time when the present embodiment uses four sub-regions to process different deformation levels in parallel. The horizontal axis represents the image deformation coefficient, and the larger the coefficient, the greater the image deformation; the vertical axis represents the average number of pixels processed per unit time, and the larger the number, the higher the processing efficiency. By comparison, it is found that the processing speed of the present embodiment is significantly improved, and the larger the image deformation coefficient, the greater the speed improvement of the method of the present embodiment.

[0088] In step S30, the step of obtaining the number of closed areas of the second image may specifically include: detecting the edge of the second image; detecting the closedness of the edge, and obtaining the number of closed areas at the same time.

[0089] Among them, detecting the edge of the second image can enhance the attachment and make the particles in the image more obvious, so that foreign matter or granular objects can be identified. Edge detection uses a first-order method and introduces a gradient operator to calculate the center difference of the pixels of the second image.

[0090] In this embodiment, the step of detecting the edge of the second image may specifically include: calculating the gradient of the second image; processing the second image using a template matrix; and calculating the final gradient amplitude and gradient direction.

[0091] The following calculation method can be used to calculate the gradient of the second image:

[0092] ,y ,

[0093] ,y .

[0094] In the above two equations, x represents the horizontal axis coordinate of the second image, y represents the vertical axis coordinate of the second image, and L represents the pixel grayscale value of the corresponding coordinate; represents the x-direction gradient; Represents the gradient in the y direction.

[0095] Processing the second image using the template matrix includes:

[0096]

[0097]

[0098] The final gradient amplitude can be calculated by the following formula:

[0099]

[0100] The gradient direction estimate is .

[0101] In this embodiment, after the edge processing is enhanced, the attachment is marked by the closure detection algorithm. The area with closed edges is marked according to a certain threshold value, and the area is used as a condition for the presence of foreign matter or particles. The number of closed areas is counted, and when the number of closed areas exceeds the first threshold value, it is determined that there are too many particles, and a signal can be output for processing by an external system.

[0102] like Figure 6As shown, it is a specific embodiment of the closure detection process. In this embodiment, detecting the closure of the edge includes the following steps:

[0103] Step S201: Start.

[0104] Step S202: Scanning the overall second image pixel information.

[0105] Step S203: Take the nth marked edge as the starting point (n=1, 2, ...).

[0106] Step S204: Acquire edge feature information.

[0107] Step S205: Use the adjacent pixel prediction method to make a prediction and establish a hash table.

[0108] Step S206: Calculate a hash value for each closed segment near the edge.

[0109] Step S207: Determine the approximation of the closed hash values ​​of adjacent pixels.

[0110] Step S208: Determine continuity using edge features.

[0111] Step S209: Record the continuous edges in a closed record list.

[0112] Step S210: Determine whether the path is closed.

[0113] Step S211: If the path is closed, update the hash table.

[0114] Step S212: If the path is not closed, determine whether to process the next marked edge; if the next marked edge is processed, return to step S204.

[0115] Step S213: If the next marked edge is not to be processed, the process ends.

[0116] Step S214: After step S211, determine whether the hash value is better than the original table; if so, return to step S206; if not, return to step S207.

[0117] In this embodiment, the closure of the edge is detected and the number of closed areas is recorded. If the number of closed areas exceeds the first threshold, it is determined that there are too many particles and the protective shell 10 needs to be cleaned or replaced, and a signal can be output for processing by an external system.

[0118] In step S30, the step of obtaining the defect coefficient of the second image may specifically include:

[0119] Step S301: Obtain the directional derivatives of all pixels of the bright field imaging image ;

[0120] Step S302: Obtain the directional derivatives of all pixels of the dark field imaging image ;

[0121] Step S303: Calculate the defect coefficient of the second image .

[0122] The order of steps S301 and S302 is not required. Obtaining the bright and dark field transmission images of the protective shell 10 can enhance the defect target, thereby facilitating the detection of defects. Bright field imaging is to allow the transmitted light beam to pass through the objective lens aperture at the back focal plane of the objective lens and block the diffraction beam to obtain the pattern contrast BFI. Dark field imaging is to tilt the incident beam direction by 2 Angle, so that the diffraction beam passes through the objective lens aperture and blocks the transmitted beam to obtain the pattern contrast DFI. Applying the bright and dark field misalignment to the same surface can obtain the defect coefficient .

[0123] Assuming that the defect coefficient of the protective shell 10 when qualified is a%, the second threshold is the sum of a% and the tolerance. The tolerance can be 10%, that is, the second threshold is (a+10)%.

[0124] like Figure 7 As shown, in one embodiment, the defect coefficient of the second image needs to be obtained through the lighting components arranged on both sides of the acquisition module 60. The lighting component may include a light source, a concave lens 6, a convex lens 7 and a baffle 8 arranged in sequence, wherein the light source is located on the side away from the protective shell 10, and the baffle 8 is located on the side close to the protective shell 10. The baffle 8 can move in the direction perpendicular to the light. The baffle 8 is provided with a slit, and the slit is used for light to pass through. The centers of the concave lens 6 and the convex lens 7 are aligned. When the slit moves to be coaxial with the vertical axis of the lens, the light source shines on the curved surface through the two lenses and the slit, and the image collected by the acquisition module 60 is a bright field image. When the slit moves to be coaxial with 1 / 2 of the distance of the vertical axis of the lens (that is, 1 / 2 of the distance between the axis and the edge of the lens), the light source shines on the curved surface through the two lenses and the slit, and the image collected by the acquisition module 60 is a dark field image.

[0125] See also Figure 8 This embodiment also provides a detection device for the protective shell 10 . The protective shell 10 is arc-shaped, and the detection device includes a collection module 60 , a processing module 2 , an acquisition module 3 and a judgment module 40 .

[0126] Among them, the acquisition module 60 is used to acquire a first image of the protective shell 10. The processing module 2 is used to perform image enhancement processing on the first image to obtain a second image. The acquisition module 3 is used to obtain the number of closed areas and / or the defect coefficient of the second image. The judgment module 40 is used to judge whether the number of closed areas exceeds the first threshold, and / or, to judge whether the defect coefficient exceeds the second threshold; if the number of closed areas exceeds the first threshold and / or the defect coefficient exceeds the second threshold, the protective shell 10 does not meet the requirements; otherwise, the protective shell 10 meets the requirements. The processing module 2 is electrically connected to the acquisition module 60 and the acquisition module 3 respectively, and the judgment module 40 is electrically connected to the acquisition module 3.

[0127] The implementation method of the detection device corresponds to the implementation method of the detection method. It can solve the technical problems solved by the implementation method of the detection method and correspondingly achieve the technical effects of the implementation method of the detection method. The specific details of this application will not be repeated here.

[0128] In this embodiment, the acquisition module 60 can be a device such as a camera or a video camera. The processing module 2 performs image enhancement processing on the first image, which belongs to pre-preprocessing. The acquisition module 3 obtains the number of closed areas and / or defect coefficients of the second image, which belongs to post-preprocessing. After the judgment module 40 obtains the result, if the protective shell 10 meets the requirements, the protective shell 10 can continue to work, and the detection method provided in this embodiment can be executed again after a predetermined interval; if the protective shell 10 does not meet the requirements, the protective shell 10 needs to be cleaned, repaired or replaced. Specifically, the acquisition module 60, the processing module 2, the acquisition module 3 and the judgment module 40 can all be electrically connected to the control component 5 to facilitate the smooth control of the detection method as a whole.

[0129] See also Figure 9-13 One embodiment of the present disclosure further provides an underwater robot, which includes a camera and a protective shell for protecting the camera. The underwater robot also includes a detection device (detection member 4) for detecting contamination of the protective shell. The detection device can detect contamination of the protective shell using the detection method in the above embodiment. The underwater robot also includes a cleaning device for cleaning the protective shell when the protective shell does not meet the requirements.

[0130] Specifically, the cleaning device includes: a cleaning device 103 of a camera 104 of an underwater robot 100 , the underwater robot 100 includes a camera 104 , and the cleaning device 103 includes a protective shell 10 , a cleaning component, a driving component, a detection component 4 and a control component 5 .

[0131] The underwater robot 100 has a front end 101 and a rear end 102 along the advancing direction X. Since the camera 104 is generally disposed at the front end 101 of the underwater robot 100, the protective shell 10 is also correspondingly located at the front end 101. Of course, in other embodiments, the camera 104 may be located at the rear end 102, the side, or other positions of the underwater robot 100, and the protective shell 10 may be correspondingly located at the rear end 102, the side, or other positions, as long as the protective shell 10 is sealed and connected to the underwater robot 100.

[0132] A pan / tilt may be provided in the protective shell 10 to control the camera 104, so that the camera 104 is more stable. The protective shell 10 is hemispherical, and the edge of the protective shell 10 is sealed and connected to the underwater robot 100. The material of the protective shell 10 may be tempered glass, or other transparent materials with a certain hardness.

[0133] like Figures 10 to 13 As shown, the cleaning assembly may include a first cleaning arm 21. The first cleaning arm 21 is arc-shaped, and both ends of the first cleaning arm 21 are located at the edge of the protective shell 10, and the middle part of the first cleaning arm 21 is bent toward the center of the protective shell 10. When the first cleaning arm 21 is working, its arc-shaped structure can form a downward channel with the outer surface 11 of the protective shell 10, which is conducive to the aquatic organisms or large attachments being pushed away from the fuselage 105, and can avoid the first cleaning arm 21 itself being attached by attachments and causing secondary pollution.

[0134] like Fig.13 As shown, the outer surface 11 of the protective shell 10 surrounded by the first cleaning arm 21 and the edge of the protective shell 10 may also be provided with a cleaning arm 23, which can increase the cleaning effect once and further clean the part close to the edge of the protective shell 10, which is more difficult to clean than the center of the protective shell 10. One end of the cleaning arm 23 is located at the edge of the protective shell 10 and between the two ends of the first cleaning arm 21, and the other end extends to the middle part of the first cleaning arm 21 and keeps a predetermined distance from the middle part, so that dirt can be discharged from the gap.

[0135] Further, the cleaning assembly may also include a second cleaning arm 22. The second cleaning arm 22 extends from the edge of the protective shell 10 to the center of the protective shell 10, extending to the center of the protective shell 10 or beyond the center of the protective shell 10. That is, one end of the second cleaning arm 22 is located at the edge of the protective shell 10, and the other end is located at the center of the protective shell 10 or beyond the center of the protective shell 10. This shape is conducive to reducing resistance. The second cleaning arm 22 is mainly used to remove dirt near the center of the hemispherical protective shell 10. Preferably, the other end of the second cleaning arm 22 is located at the center of the protective shell 10, that is, at the apex of the hemisphere, so that after the protective shell rotates around its center for one circle, the path passed by the second cleaning arm 22 can cover the entire outer surface 11 of the protective shell 10, thereby making the cleaning more comprehensive.

[0136] In a specific embodiment, the cleaning assembly may include two first cleaning arms 21 and one second cleaning arm 22. One end of the second cleaning arm 22 is located at the lowest point of the edge of the protective shell 10, and the two first cleaning arms 21 are symmetrically distributed about the line connecting the second cleaning arm 22 and the center of the protective shell 10. Preferably, the two first cleaning arms 21 have the same shape. The two first cleaning arms 21 may be located at the leftmost and rightmost ends of the edge of the protective shell 10, respectively. Alternatively, the two first cleaning arms 21 and one second cleaning arm 22 are respectively located at the three vertices of an equilateral triangle, so that the overall distribution of the cleaning assembly is more uniform, thereby achieving a better cleaning effect.

[0137] The middle part of the first cleaning arm 21 does not reach the center of the protective shell 10, so that the two first cleaning arms 21 will not collide. In this embodiment, the camera 104 is close to the highest point of the edge of the protective shell 10, so that the second cleaning arm 22 will not block the shooting range.

[0138] In this embodiment, one end of the cleaning component, the detection component 4 and the control component 5 can be fixed to the body 105 of the underwater robot 100. Specifically, the cleaning component is located at the edge of the protective shell 10 and is fixed to the body 105 of the underwater robot 100. Figure 1 As shown, the detection component 4 and the control component 5 can be fixed in the fuselage 105, and the detection component 4 and the control component 5 can be arranged along the forward direction X of the underwater robot 100, and the detection component 4 and the control component 5 are both higher than the lowest point of the protective shell 10. Fig.10 As shown, the detection component 4 and the control component 5 can be fixed outside the fuselage 105 , and the detection component 4 and the control component 5 can be arranged perpendicular to the forward direction X of the underwater robot 100 , and the detection component 4 and the control component 5 are both located below the protective shell 10 .

[0139] like Fig. 9 , Fig.10 and Fig.11 As shown, the detection member 4 is located between the control member 5 and the protective shell 10. The detection member 4 can determine whether the dirtiness of the protective shell 10 exceeds a threshold value by image recognition. The detection range of the detection member 4 can coincide with the shooting range / visual range of the camera 104. Of course, in other embodiments, it can also be larger or slightly smaller than the shooting range / visual range. The detection range of the detection member 4 mainly includes the upper half of the protective shell 10, because the camera 104 is located in the upper half of the inner side of the protective shell 10. Figure 1 and Figure 3 As shown, the detection range of the detection element 4 is between the two dotted lines in the figure.

[0140] The driving assembly may include a motor 31 and a first gear 32. The motor 31 includes a driving shaft 33, and the driving shaft 33 is used to drive the first gear 32 to rotate. A second gear is provided on the protective shell 10, and the first gear 32 and the second gear are meshed and connected, so that the motor 31 can drive the first gear 32 and the second gear to rotate through the driving shaft 33 to realize the rotation of the protective shell 10. Preferably, the center of the protective shell 10 is stationary, that is, the protective shell 10 rotates around its center. Fig.12 As shown, the dotted arrow indicates the rotation direction of the protective shell 10, which can rotate counterclockwise, and can also rotate clockwise in other embodiments.

[0141] Specifically, the detection component 4 and the control component 5 can be connected via a wired communication cable to achieve two-way communication, and can use protocols such as Ethernet and RS485. The detection component 4 sends the detected dirtiness image or value to the control component 5 via the cable, and the control component 5 determines whether the dirtiness exceeds a threshold value. If it exceeds the threshold value, the control component 5 sends a start signal to the motor 31, and controls the motor 31 to start working via the cable.

[0142] In a specific application scenario, the detection member 4 detects the degree of dirtiness of the outer surface 11 of the protective shell 10. When the degree of dirtiness exceeds the threshold, the control member 5 controls the drive assembly to start, so that the protective shell 10 starts to rotate. During rotation, the outer surface 11 of the protective shell 10 will pass through the first cleaning arm 21, the second cleaning arm 22, and the first cleaning arm 21 in sequence. The first first cleaning arm 21 is responsible for removing larger attachments attached to the outer surface 11 of the protective shell 10, such as underwater organisms, large particles of attachments, etc.; the second cleaning arm 22 is responsible for removing attachments of medium-sized particles; the second first cleaning arm 21 is responsible for removing smaller particles of attachments. Of course, the order of contact between dirt and each cleaning arm may be different, and this application does not make the sole limitation on this. After the protective shell 10 rotates for one cycle, the control member 5 returns the command to the detection member 4, and the detection member 4 executes the detection program again. If the degree of dirtiness does not exceed the threshold, it stops working, otherwise it performs the cleaning operation again until the conditions are met. One of the cycles can be that the protective shell 10 rotates 360°, or other required angles.

[0143] An embodiment of the present disclosure also provides a method for cleaning an underwater robot camera, the cleaning method adopts the cleaning device as described above, and the cleaning method includes the following steps: using the detection member to detect the degree of dirtiness of the outer surface of the protective shell; judging whether the degree of dirtiness exceeds the threshold value; when the degree of dirtiness exceeds the threshold value, the control member controls the drive component to start; when the degree of dirtiness does not exceed the threshold value, performing the step of using the detection member to detect the degree of dirtiness of the protective shell again at an interval of a predetermined time; after the drive component is started, the protective shell is driven to rotate a predetermined angle, and at the same time, the cleaning component cleans the outer surface of the protective shell; after the protective shell rotates the predetermined angle, returning to the step of using the detection member to detect the degree of dirtiness of the protective shell.

[0144] It should be noted that, in the description of this specification, the terms "first", "second", etc. are only used for descriptive purposes and to distinguish similar objects. There is no order of precedence between the two, and they cannot be understood as indicating or implying relative importance. In addition, in the description of this specification, unless otherwise specified, the meaning of "plurality" is two or more.

[0145] The use of the term "comprising" or "including" to describe the combination of elements, ingredients, parts or steps herein also contemplates an embodiment that is basically composed of these elements, ingredients, parts or steps. By using the term "may" here, it is intended to illustrate that any attribute described that "may" include is optional.

[0146] Multiple elements, ingredients, parts or steps can be provided by a single integrated element, ingredient, part or step. Alternatively, a single integrated element, ingredient, part or step can be divided into separate multiple elements, ingredients, parts or steps. The disclosure "one" or "an" used to describe an element, ingredient, part or step is not intended to exclude other elements, ingredients, parts or steps.

[0147] It should be understood that the above description is for illustration and not for limitation. Many embodiments and many applications beyond the examples provided will be apparent to those skilled in the art upon reading the above description. Therefore, the scope of the present teachings should not be determined with reference to the above description, but should be determined with reference to the appended claims and the full scope of equivalents possessed by such claims. For the purpose of comprehensiveness, all articles and references, including disclosures of patent applications and publications, are incorporated herein by reference. The omission of any aspect of the subject matter disclosed herein in the preceding claims is not intended to be a waiver of the subject matter, nor should it be considered that the inventors did not consider the subject matter to be part of the disclosed inventive subject matter.

Claims

1. A method for detecting a protective shell of an underwater robot camera, characterized in that: The protective shell is arc-shaped, and the detection method comprises the following steps: capturing a first image of the protective shell; Performing image enhancement processing on the first image to obtain a second image; Acquiring the number of closed areas and / or defect coefficient of the second image; Determining whether the number of closed areas exceeds a first threshold, and / or determining whether the defect coefficient exceeds a second threshold; If the number of closed areas exceeds the first threshold and / or the defect coefficient exceeds the second threshold, the protective shell does not meet the requirements; Otherwise, the protective shell meets the requirements; The protective shell has a detection area, and the detection area includes a plurality of sub-areas; In the step of acquiring the first image of the protective shell, the first image of each of the sub-areas is acquired simultaneously; In the step of performing image enhancement processing on the first image, image enhancement processing is performed on the first images of the sub-regions simultaneously and in parallel; The step of performing image enhancement processing on the first image to obtain the second image comprises: Processing the first image with a correction algorithm to obtain a corrected image; Performing interpolation processing on the gaps of the corrected image to obtain the second image; The step of obtaining the number of closed areas of the second image comprises: detecting edges of the second image; Detecting the closure of the edge and obtaining the number of the closed areas; The step of detecting the edge of the second image comprises: calculating a gradient of the second image; Processing the second image using a template matrix; Calculate the final gradient magnitude and gradient direction; The step of obtaining the defect coefficient of the second image comprises: Get the directional derivatives of all pixels in the bright field image ; Get the directional derivatives of all pixels in the dark field imaging image ; Calculate the defect coefficient of the second image .

2. The method for detecting the protective shell of the underwater robot camera according to claim 1, characterized in that: The protective shell is hemispherical, and the diameter of the protective shell is 20 cm to 200 cm.

3. The method for detecting the protective shell of the underwater robot camera according to claim 1, characterized in that: The detection area includes four rectangular sub-areas, and each of the sub-areas has the same width and length.

4. The method for detecting the protective shell of the underwater robot camera according to claim 1, characterized in that: The image enhancement process includes correction, filtering, quantization, and white balance.

5. An underwater robot, characterized in that: The underwater robot comprises a camera and a protective shell for protecting the camera, wherein the protective shell is arc-shaped, and the underwater robot further comprises a detection device for detecting contamination of the protective shell; The protective shell has a detection area, and the detection area includes a plurality of sub-areas; wherein the detection device includes: A collection module, used to collect the first image of the protective shell and the first image of each sub-area at the same time; A processing module, configured to perform image enhancement processing on the first image to obtain a second image, and simultaneously perform image enhancement processing on the first image of each of the sub-regions in parallel; An acquisition module, used for acquiring the number of closed areas and / or defect coefficient of the second image; a judgment module, used to judge whether the number of closed areas exceeds a first threshold value, and / or whether the defect coefficient exceeds a second threshold value; if the number of closed areas exceeds the first threshold value and / or the defect coefficient exceeds the second threshold value, the protective shell does not meet the requirements; otherwise, the protective shell meets the requirements; the underwater robot also includes a cleaning device for cleaning the protective shell when the protective shell does not meet the requirements; Wherein, the processing module is electrically connected to the collection module and the acquisition module respectively, and the judgment module is electrically connected to the acquisition module; The obtaining of the second image comprises: Processing the first image with a correction algorithm to obtain a corrected image; Performing interpolation processing on the gaps of the corrected image to obtain the second image; The acquiring the number of closed areas of the second image comprises: detecting edges of the second image; Detecting the closure of the edge and obtaining the number of the closed areas; Detecting the edge of the second image comprises: calculating a gradient of the second image; Processing the second image using a template matrix; Calculate the final gradient magnitude and gradient direction; The acquiring the defect coefficient of the second image comprises: Get the directional derivatives of all pixels in the bright field image ; Get the directional derivatives of all pixels in the dark field imaging image ; Calculate the defect coefficient of the second image .

Images