Image processing method and system suitable for hydraulic tunnel detection

Abstract

The invention provides an image processing method and system suitable for hydraulic tunnel detection. According to the implementation scheme, based on vault area image contents in two adjacent tunnel section images, vault area curvature features are determined so as to identify an irregular settlement area in a corresponding tunnel surface area, and the curvature features of the area are obtained; and optimizing and adjusting a view field overlapping area between two adjacent tunnel section images by utilizing the curvature characteristics of the irregular settlement area. Based on the adjusted view field overlapping area, two feature point sets are extracted from two adjacent tunnel section images respectively, if the matching degree of the two sets meets the requirement, the two sets are fused to obtain a fused feature point set, and the fused feature point set is used as a fused feature point set; and splicing two adjacent tunnel section images by using the fused feature point set to obtain a tunnel spliced image for deformation detection of the tunnel spliced image. According to the invention, the detection accuracy of tunnel deformation can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, and in particular to an image processing method and system suitable for water tunnel detection. BACKGROUND

[0002] Water tunnels play an important role in water conservancy projects in China and are important infrastructure for water conveyance, irrigation and other projects. After a long period of operation, a large number of water tunnels have cracks, damage, corrosion, leakage, landslides and other diseases. If these defects are not promptly identified and addressed, they will inevitably affect the safe operation of the water tunnel, shorten the service life of the tunnel, and even cause safety accidents in the tunnel, resulting in significant economic losses and even casualties.

[0003] Due to the harsh conditions of water tunnels, such as wet environments with accumulated water in the tunnel, no lighting, and only manual access conditions, traditional highway, subway and other vehicle-mounted tunnel detection equipment and methods cannot be applied to water tunnel detection. Manual detection methods are slow, time-consuming and ineffective, cannot form a benchmark database, and it is difficult to compare multiple exploration data, resulting in problems such as unclear disease causes, unclear disease information, unclear disease progression, and unclear health status. Tunnel disease detection methods urgently need to be upgraded.

[0004] Therefore, in some technologies, cross-sectional images of the tunnel are collected, and each cross-sectional image is spliced to obtain a complete tunnel image. Then, the tunnel image is recognized in an artificial intelligence manner to obtain a tunnel detection result. In this way, the detection accuracy and efficiency can be improved.

[0005] However, in order to ensure the splicing quality, there is usually a certain overlap area between adjacent images. Setting these overlap areas to improve the quality of the spliced image is a technical problem to be solved in the field. SUMMARY

[0006] The present application provides an image processing method and system suitable for water tunnel detection, which can solve at least one of the above technical problems.

[0007] According to an aspect of the present application, an image processing method suitable for water tunnel detection is provided, comprising: determining a vault region curvature feature based on a vault region image in a first tunnel cross-sectional image and a vault region image in a second tunnel cross-sectional image in two adjacent tunnel cross-sectional images, wherein there is a field of view overlap region between the two adjacent tunnel cross-sectional images; determining an irregular settlement region in a tunnel surface region corresponding to the two adjacent tunnel cross-sectional images based on the vault region curvature feature, and a curvature feature of the irregular settlement region; adjust the field of view overlap region between the two adjacent tunnel section images based on the curvature feature of the irregular settlement region; extract a first feature point set from the first tunnel section image and a second feature point set from the second tunnel section image based on the adjusted field of view overlap region; if the matching degree between the first feature point set and the second feature point set meets a preset first matching condition, perform image stitching on the first tunnel section image and the second tunnel section image based on the first feature point set and the second feature point set to obtain a tunnel stitching image.

[0008] According to another aspect of the present application, an image processing device suitable for water tunnel detection is provided, comprising: a first feature determination module configured to determine a vault region curvature feature based on a vault region image in a first tunnel section image and a vault region image in a second tunnel section image of two adjacent tunnel section images, wherein there is a field of view overlap region between the two adjacent tunnel section images; a second feature determination module configured to determine an irregular settlement region in a tunnel surface region corresponding to the two adjacent tunnel section images and a curvature feature of the irregular settlement region based on the vault region curvature feature; an overlap region adjustment module configured to adjust the field of view overlap region between the two adjacent tunnel section images based on the curvature feature of the irregular settlement region; a feature point extraction module configured to extract a first feature point set from the first tunnel section image and a second feature point set from the second tunnel section image based on the adjusted field of view overlap region; an image stitching module configured to, if the matching degree between the first feature point set and the second feature point set meets a preset first matching condition, perform image stitching on the first tunnel section image and the second tunnel section image based on the first feature point set and the second feature point set to obtain a tunnel stitching image.

[0009] According to another aspect of the present application, an image processing system suitable for water tunnel detection is provided, comprising at least one processor and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the processor is configured to acquire the instructions from the memory and execute the instructions to enable the at least one processor to execute the image processing method suitable for water tunnel detection according to any of the embodiments of the present application.

[0010] According to another aspect of the present application, there is provided a non-transitory computer readable storage medium storing computer instructions for providing to a computer to instruct the computer to perform the image processing method for water tunnel detection according to any of the embodiments of the present application.

[0011] By using the technical scheme of the present application, the curvature feature of the crown region is determined by the image content of the crown region in the two adjacent tunnel section images, so that the irregular settlement region in the tunnel surface region corresponding to the two adjacent tunnel section images can be accurately identified by using the feature, and the curvature feature of the region is obtained. The visual field overlapping region between the two adjacent tunnel section images is adjusted and optimized by using the curvature feature of the irregular settlement region. In this way, based on the adjusted visual field overlapping region, two feature point sets are extracted from the two adjacent tunnel section images respectively. If the matching degree of the two feature point sets meets the requirement, the two feature point sets are fused to obtain a fused feature point set, and the two adjacent tunnel section images are spliced by using the fused feature point set to obtain a tunnel splicing image. In this way, the overlapping region is adjusted by identifying the curvature feature of the irregular settlement region, and the tunnel splicing image is optimized. Furthermore, the deformation detection accuracy of the tunnel splicing image can be improved.

[0012] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present application, nor to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0013] The accompanying drawings are used to better understand the present application, and do not constitute a limitation of the present application. Among them: Figure 1 is a flowchart of the image processing method for water tunnel detection according to an embodiment of the present application; Figure 2 is a schematic diagram of a handheld device according to an embodiment of the present application; Figure 3 is a schematic diagram of a tunnel according to an embodiment of the present application; Figure 4 is a schematic diagram of a tunnel splicing image according to an embodiment of the present application; Figure 5 is a structural block diagram of an image processing device for water tunnel detection according to an embodiment of the present application; Figure 6 is a block diagram of an electronic device for implementing the method according to an embodiment of the present application. DETAILED DESCRIPTION

[0014] Exemplary embodiments of the present application are described herein with reference to the accompanying drawings, which are meant to be exemplary in nature, and include various specific details to facilitate understanding, and are to be considered in conjunction with the description. Thus, one of ordinary skill in the art will appreciate that various changes and modifications can be made to the embodiments described herein without departing from the scope of the present application. Likewise, the description herein is meant to be illustrative only and is not meant to be limiting in any way.

[0015] Figure 1 is a flowchart of an image processing method suitable for water tunnel detection according to an embodiment of the present application.

[0016] As shown in Figure 1 , the image processing method suitable for water tunnel detection can comprise: S110, determining a vault region curvature feature based on a vault region image in a first tunnel section image and a vault region image in a second tunnel section image in the two adjacent tunnel section images, wherein there is a field of view overlap region between the two adjacent tunnel section images; S120, determining an irregular settlement region in a tunnel surface region corresponding to the two adjacent tunnel section images and a curvature feature of the irregular settlement region based on the vault region curvature feature; S130, adjusting the field of view overlap region between the two adjacent tunnel section images based on the curvature feature of the irregular settlement region; S140, extracting a first feature point set from the first tunnel section image and a second feature point set from the second tunnel section image based on the adjusted field of view overlap region; S150, in the case that the matching degree between the first feature point set and the second feature point set meets a preset first matching condition, performing image stitching on the first tunnel section image and the second tunnel section image based on the first feature point set and the second feature point set to obtain a tunnel stitching image.

[0017] It can be understood that the first tunnel section image can be the first tunnel stitching image photographed first or the tunnel stitching image stitched from the tunnel section images photographed before. The second tunnel section image can be the tunnel section image currently photographed. In this way, a complete tunnel image can be obtained after multiple stitching.

[0018] Exemplarily, after obtaining the tunnel stitching image, deformation detection can be performed on the tunnel stitching image to obtain tunnel deformation information of the tunnel surface region. For example, a trained neural network model can be used to perform deformation detection on the tunnel stitching image to obtain the tunnel deformation information of the tunnel surface region.

[0019] Exemplarily, the embodiments of the present application can be applied toFigure 2 The handheld device can collect images of the tunnel, generate a tunnel splicing image, and perform deformation detection on the tunnel splicing image to output tunnel deformation information of a tunnel surface region. For example, the tunnel deformation information can include information such as the location, size, and the like of a crack, a hole, or exposed steel bars. Figure 3 The handheld device can collect images of the tunnel, generate a tunnel splicing image, and perform deformation detection on the tunnel splicing image to output tunnel deformation information of a tunnel surface region. For example, the tunnel deformation information can include information such as the location, size, and the like of a crack, a hole, or exposed steel bars.

[0020] For example, the control accuracy of the motor in the handheld device is not less than 0.001°. When a tunnel picture is taken at a certain section, the pose control accuracy of the camera is 0.001°. Thus, in the later image splicing optimization, each picture can obtain a relatively ideal initial value. The monocular rotation angle can be artificially set, and the coincidence degree of every two pictures is preferably higher than 30°.

[0021] For example, the first tunnel section image and the second tunnel section image can be images of the inner surface of the tunnel ring. There is an overlapping field of view between the two images, that is, the image content of the front and rear connection positions between the two images is substantially the same, which facilitates improving the image splicing quality.

[0022] For example, the first tunnel section image and the second tunnel section image can be point cloud images.

[0023] For example, the basic image collection operation of the handheld device can be as follows: The measurement personnel holds the device, and the device can be connected to the measurement personnel's mobile phone through Bluetooth. The measurement personnel presses a photographing button at one of the collection points, and the device takes a ring-shaped picture of the tunnel section at the collection point. The measurement personnel moves to another collection point, presses the photographing button at the other collection point, and takes a ring-shaped picture of the next section. In this way, two adjacent tunnel section images are obtained. The measurement personnel collects tunnel section images at each collection point in this way, and at the last collection point, a finish button is pressed, and all the pictures are spliced in the manner of the embodiment of the present application, and the finally spliced tunnel splicing image is stored in an SD card.

[0024] As shown in Figure 4 The handheld device can be connected to a background server, the background server analyzes the tunnel splicing image to label potential deformation conditions such as cracks, bulges, and exposed steel bars, and labels the location and size of the deformation conditions. In this way, the measurement personnel can manage (review, comment, modify, and the like) the tunnel splicing image in the later stage, and compare the differences between tunnel splicing images collected at different times.

[0025] Exemplarily, the vault region curvature feature can include curvature, curvature change rate, curvature change direction, etc., or the vault region curvature feature can also include a distribution map of the curvature with respect to position. The distribution map includes the curvature at each tunnel inner surface position.

[0026] It can be understood that the vault region is the inner surface region of the tunnel vault.

[0027] Exemplarily, the vault region curvature feature is extracted from the vault region image in the first tunnel section image, and the vault region curvature feature is extracted from the vault region image in the second tunnel section image, and then the two features are combined to obtain the final vault region curvature feature.

[0028] Exemplarily, the irregular settlement region can be understood as follows: the tunnel plane at a first position in the tunnel is settled relative to the tunnel plane at a second position, the first position is the settlement region, and the shape of the settlement region is irregular.

[0029] Exemplarily, the curvature feature of the regular settlement region can include curvature, curvature change rate, curvature change direction, etc., or the vault region curvature feature can also include a distribution map of the curvature with respect to position.

[0030] Exemplarily, if the curvature of a target position in the irregular settlement region is too large, and the target position falls on the edge of the field of view overlap region, the target position can be expanded in the field of view overlap region. For example, a region growing algorithm can be used to generate point clouds of the target position in the field of view overlap region in the two images to reach a set growing range, so as to obtain an adjusted field of view overlap region.

[0031] Exemplarily, from the first tunnel section image, feature points in the adjusted field of view overlap region are extracted to obtain a first feature point set. From the second tunnel section image, feature points in the adjusted field of view overlap region are extracted to obtain a second feature point set. The first feature point set can include a plurality of feature point clouds. The second feature point set can include a plurality of feature point clouds.

[0032] Exemplarily, the union of the first feature point set and the second feature point set is obtained to obtain a third feature point set. Alternatively, part of the features in the first feature point set and the second feature point set are fused or removed, etc., to obtain the third feature point set.

[0033] Exemplarily, based on the third feature point set, the relative pose between the first tunnel section image and the second tunnel section image is determined, and the image relative spatial relationship represented by the relative pose is used to perform image stitching on the first tunnel section image and the second tunnel section image to obtain a tunnel stitching image.

[0034] Exemplarily, in a case that the matching degree between the first set of feature points and the second set of feature points meets a preset first matching condition, the second tunnel section image can be rephotographed. Alternatively, the field of view overlap region is re-adjusted by using step S130 until the matching degree between the first set of feature points and the second set of feature points meets the first matching condition.

[0035] Exemplarily, the first matching condition can be that the matching degree between the first set of feature points and the second set of feature points is greater than a preset matching degree threshold.

[0036] According to the above embodiment, the vault region curvature feature is determined by the image content of the vault region in the two adjacent tunnel section images, so that the irregular settlement region in the tunnel surface region corresponding to the two adjacent tunnel section images can be accurately identified by using the feature, and the curvature feature of the region is obtained. The field of view overlap region between the two adjacent tunnel section images is adjusted by using the curvature feature of the irregular settlement region. In this way, two sets of feature points are extracted from the two adjacent tunnel section images based on the adjusted field of view overlap region, and if the matching degree of the two sets of feature points meets the requirement, the two sets of feature points are fused to obtain a fused set of feature points, and the two adjacent tunnel section images are spliced by using the fused set of feature points to obtain a tunnel spliced image. In this way, the overlap region is adjusted by identifying the curvature feature of the irregular settlement region, and then the tunnel spliced image is optimized. Further, the deformation detection accuracy of the tunnel spliced image can be improved.

[0037] In an embodiment, the vault region curvature feature is determined based on the vault region image in the first tunnel section image and the vault region image in the second tunnel section image in the two adjacent tunnel section images, comprising: extracting a plurality of first vault profile line segments from the vault region image in the first tunnel section image, and performing circular arc curve fitting on each first vault profile line segment to obtain each first circular arc curve; determining a first vault region curvature distribution map based on the curvature corresponding to each first circular arc curve and the position of each first circular arc curve in the vault region image; extracting a plurality of second vault profile line segments from the vault region image in the second tunnel section image, and performing circular arc curve fitting on each second vault profile line segment to obtain each second circular arc curve; determining a second vault region curvature distribution map based on the curvature corresponding to each second circular arc curve and the position of each second circular arc curve in the vault region image; and determining the vault region curvature feature based on the first vault region curvature distribution map and the second vault region curvature distribution map.

[0038] Exemplarily, the tunnel section image includes a vault region image and a non-vault region image.

[0039] Exemplarily, a Gaussian filter can be employed to denoise the vault region image, and then the vault contour line segment can be extracted from the denoised vault region image. For example, an edge detection algorithm is employed to process the denoised vault region image, and a plurality of first vault contour line segments are obtained.

[0040] Exemplarily, the vault contour line segment includes a plurality of point clouds. A least square method or a polynomial function is employed to fit the point clouds in the vault contour line segment, and a circular arc curve is obtained.

[0041] Exemplarily, the vault region curvature distribution map is used to describe the curvature changing with the position.

[0042] Exemplarily, the first vault region curvature distribution map and the second vault region curvature distribution map are merged to obtain a third vault region curvature distribution map, and the third vault region curvature distribution map is taken as the vault region curvature feature.

[0043] According to the above embodiment, by fitting the vault contour line segments in the adjacent two tunnel section images, the corresponding circular arc curves can be obtained, and by using the curvatures of the circular arc curves and the positions of the circular arc curves, the vault region curvature distribution map can be obtained. In this way, the vault region curvature distribution maps extracted from the two tunnel section images are merged, and the curvature feature of the vault region in the tunnel surface region can be accurately described.

[0044] In an embodiment, based on the vault region curvature feature, an irregular settlement region in a tunnel surface region corresponding to the adjacent two tunnel section images is determined, and a curvature feature of the irregular settlement region includes: based on the vault region curvature feature, a third vault region curvature distribution map is determined; in the third vault region curvature distribution map, a vault region curvature change rate and a curvature change direction in each continuous sampling interval are determined; in a case where the vault region curvature change rate in a first sampling interval in each sampling interval is greater than a preset change rate threshold, and the curvature change direction meets a preset direction change condition, the first sampling interval is determined as the irregular settlement region, and based on the vault region curvature change rate and the curvature change direction of the first sampling interval, the curvature feature of the irregular settlement region is determined.

[0045] Exemplarily, a plurality of sampling intervals can be set in the third vault region curvature distribution map, and each sampling interval can be continuous.

[0046] Exemplarily, if the vault region curvature change rate in the first sampling interval is greater than the preset change rate threshold, and the curvature change direction maintains monotonicity, the first sampling interval is taken as the irregular settlement region, i.e., the key region of deformation monitoring.

[0047] Exemplarily, if the curvature change rate continuously increases or continuously decreases, it is considered that the curvature change direction maintains monotonicity.

[0048] Exemplarily, there can be one or more irregular settlement areas, and each irregular settlement area can be determined in the above manner.

[0049] Exemplarily, the curvature change rate and the curvature change direction of the vault region in the first sampling interval are taken as the curvature features of the irregular settlement area.

[0050] According to the above embodiment, by evaluating the curvature change rate and the curvature change direction of the vault region in each sampling interval in the three-vault region curvature distribution diagram, it can be accurately determined whether the sampling interval is an irregular settlement area, so that the irregular settlement area can be monitored in particular.

[0051] In an embodiment, based on the curvature features of the irregular settlement area, the field of view overlap region between the adjacent two tunnel section images is adjusted, including: based on the curvature change rate of the vault region in the curvature features of the irregular settlement area, determining the curvature change amplitude of the irregular settlement area; based on the curvature change amplitude, determining the field of view overlap ratio; based on the position information of the irregular settlement area and the boundary position of the field of view overlap region between the adjacent two tunnel section images, determining the to-be-adjusted position of the field of view overlap region; based on the field of view overlap ratio, adjusting the to-be-adjusted position of the field of view overlap region to obtain an adjusted field of view overlap region.

[0052] Exemplarily, based on the difference between the minimum value and the maximum value of the curvature change rate of the vault region in the curvature features of the irregular settlement area, the curvature change amplitude of the irregular settlement area is determined.

[0053] Exemplarily, the curvature change amplitude is proportional to the field of view overlap ratio. The greater the curvature change amplitude, the greater the field of view overlap ratio.

[0054] Exemplarily, if the distance between the irregular settlement area and the nearest boundary position of the field of view overlap region is less than a preset threshold, it is determined that the adjusted field of view overlap region needs to include the irregular settlement area, that is, the nearest boundary position is taken as the to-be-adjusted position for adjustment, so that the irregular settlement area falls in the field of view overlap region. For example, a region growing algorithm can be used to generate point clouds for the nearest boundary position. For another example, the proportion between the generated field of view overlap region and the second tunnel section image needs to reach the field of view overlap ratio.

[0055] According to the above embodiment, based on the curvature characteristics of the irregular subsidence area, the field of view overlap area between the two adjacent tunnel section images is adjusted to include the irregular subsidence area, so that the image quality of this part of the area can be improved when the images are subsequently spliced, and the accuracy of the deformation detection of this area is further improved.

[0056] In an embodiment, the image splicing of the first tunnel section image and the second tunnel section image based on the first feature point set and the second feature point set to obtain a tunnel spliced image comprises: performing feature fusion on the first feature point set and the second feature point set to obtain a third feature point set; and performing image splicing on the first tunnel section image and the second tunnel section image based on the third feature point set to obtain a tunnel spliced image.

[0057] Illustratively, the union of the first feature point set and the second feature point set is obtained to obtain the third feature point set. Alternatively, the first feature point set and the second feature point set are fused or removed, etc. to obtain the third feature point set.

[0058] Illustratively, based on the third feature point set, the relative pose between the first tunnel section image and the second tunnel section image is determined, and the image splicing of the first tunnel section image and the second tunnel section image is performed using the relative spatial relationship represented by the relative pose to obtain a tunnel spliced image.

[0059] According to the above embodiment, the relative spatial position relationship between the two images can be determined using the feature point set of the overlap area of the two adjacent tunnel section images, and the two adjacent tunnel section images can be accurately spliced using the relative spatial position relationship, and the image quality of the tunnel spliced image is improved.

[0060] In an embodiment, the feature fusion of the first feature point set and the second feature point set to obtain a third feature point set comprises: in the case where the matching degree of a first feature point in the first feature point set and a second feature point in the second feature point set is greater than a second matching condition, the first feature point and the second feature point are fused to obtain a third feature point, and the third feature point is added to the third feature point set; and in the case where the matching degree of a third feature point in the first feature point set and a fourth feature point in the second feature point set is less than the second matching condition, the third feature point and the fourth feature point are added to the third feature point set.

[0061] Illustratively, the second matching condition can be that the matching degree of two feature points is greater than a preset matching degree threshold.

[0062] Exemplarily, the fusing the first feature point with the second feature point can include: averaging the two feature points to take the average feature as the third feature point.

[0063] According to the above embodiment, the feature fusion is performed on the first feature point set and the second feature point set, and the third feature point set can be accurately calculated.

[0064] In an embodiment, further comprising: in a case where the matching degree between the first feature point set and the second feature point set does not meet the first matching condition, determining an adjustment mode of the field of view overlap region based on the matching degree between the first feature point set and the second feature point set; and adjusting the field of view overlap region between the two adjacent tunnel section images again based on the adjustment mode of the field of view overlap region.

[0065] Exemplarily, if the matching degree of the two sets is lower than the preset matching degree threshold, the adjustment mode of the field of view overlap region is to re-shoot the two adjacent tunnel section images, and to perform edge detection on the two adjacent tunnel section images re-shot to obtain the adjusted field of view overlap region.

[0066] Exemplarily, if the matching degree of the two sets is higher than the preset matching degree threshold, the adjustment mode of the field of view overlap region is to use a region growing algorithm to perform region growing on the field of view overlap region to obtain the adjusted field of view overlap region.

[0067] Exemplarily, the field of view overlap region can be continuously adjusted in the above manner until the matching degree between the first feature point set and the second feature point set meets the first matching condition, and then the step S150 is executed.

[0068] According to the above embodiment, in a case where the matching degree between the two feature point sets of the field of view overlap region of the two images does not meet the preset threshold, the matching degree is used to determine the adjustment mode, and the field of view overlap region is adjusted again using the adjustment mode, so that the image splicing quality and the detection accuracy of the spliced image can be improved.

[0069] Figure 5 is a structural block diagram of an image processing device suitable for water tunnel detection according to an embodiment of the present application.

[0070] As shown in Figure 5 , the image processing device suitable for water tunnel detection can include: A first feature determination module 510 is configured to determine a vault region curvature feature based on a vault region image in a first tunnel section image and a vault region image in a second tunnel section image of the two adjacent tunnel section images, wherein the two adjacent tunnel section images have a field of view overlap region. The second feature determination module 520 is configured to determine, based on the vault region curvature feature, an irregular subsidence region in a tunnel surface region corresponding to the two adjacent tunnel section images, and a curvature feature of the irregular subsidence region. The first region adjustment module 530 is configured to adjust a field of view overlap region between the two adjacent tunnel section images based on the curvature feature of the irregular subsidence region. The feature point extraction module 540 is configured to extract a first feature point set from the first tunnel section image and a second feature point set from the second tunnel section image based on the adjusted field of view overlap region. The image stitching module 550 is configured to, when a matching degree between the first feature point set and the second feature point set meets a preset first matching condition, stitch the first tunnel section image and the second tunnel section image based on the first feature point set and the second feature point set to obtain a tunnel stitching image.

[0071] In an embodiment, the first feature determination module 510 includes: The first fitting unit is configured to extract a plurality of first vault profile line segments from a vault region image in the first tunnel section image, and perform circular arc curve fitting on each first vault profile line segment to obtain a first circular arc curve. The first distribution determination unit is configured to determine a first vault region curvature distribution map based on a curvature corresponding to each first circular arc curve and a position of each first circular arc curve in the vault region image. The second fitting unit is configured to extract a plurality of second vault profile line segments from a vault region image in the second tunnel section image, and perform circular arc curve fitting on each second vault profile line segment to obtain a second circular arc curve. The second distribution determination unit is configured to determine a second vault region curvature distribution map based on a curvature corresponding to each second circular arc curve and a position of each second circular arc curve in the vault region image. The first feature determination unit is configured to determine the vault region curvature feature based on the first vault region curvature distribution map and the second vault region curvature distribution map.

[0072] In an embodiment, the second feature determination module 520 includes: The third distribution determination unit is configured to determine a third vault region curvature distribution map based on the vault region curvature feature. The sampling feature determination unit is configured to determine, in the third vault region curvature distribution map, a vault region curvature change rate and a curvature change direction in each continuous sampling interval. The second feature determination unit is configured to determine, in a case where a curvature change rate of a vault region in a first sampling interval of the sampling intervals is greater than a preset change rate threshold and a curvature change direction meets a preset direction change condition, that the first sampling interval is an irregular settlement region, and determine a curvature feature of the irregular settlement region based on the curvature change rate and the curvature change direction of the vault region in the first sampling interval.

[0073] In an embodiment, the first region adjustment module 530 includes: The amplitude determination unit is configured to determine a curvature change amplitude of the irregular settlement region based on the curvature change rate of the vault region in the curvature feature of the irregular settlement region. The overlap ratio determination unit is configured to determine a field of view overlap ratio based on the curvature change amplitude. The adjustment position determination unit is configured to determine a to-be-adjusted position of a field of view overlap region based on position information of the irregular settlement region and a boundary position of the field of view overlap region between the two adjacent tunnel section images. The position adjustment unit is configured to adjust the to-be-adjusted position of the field of view overlap region based on the field of view overlap ratio, to obtain an adjusted field of view overlap region.

[0074] In an embodiment, the image stitching module 550 includes: The feature fusion unit is configured to perform feature fusion on the first feature point set and the second feature point set, to obtain a third feature point set. The image stitching unit is configured to perform image stitching on the first tunnel section image and the second tunnel section image based on the third feature point set, to obtain a tunnel stitching image.

[0075] In an embodiment, the feature fusion unit is specifically configured to: In a case where a matching degree of a first feature point in the first feature point set and a second feature point in the second feature point set is greater than a second matching condition, the first feature point and the second feature point are fused to obtain a third feature point, and the third feature point is added to the third feature point set. In a case where a matching degree of a third feature point in the first feature point set and a fourth feature point in the second feature point set is less than the second matching condition, the third feature point and the fourth feature point are added to the third feature point set.

[0076] In an embodiment, the apparatus further includes: The adjustment mode determination module is configured to determine an adjustment mode of the field of view overlap region based on the matching degree between the first set of feature points and the second set of feature points, in a case where the matching degree between the first set of feature points and the second set of feature points does not meet the first matching condition. The second region adjustment module is configured to adjust the field of view overlap region between the two adjacent tunnel section images again based on the adjustment mode of the field of view overlap region.

[0077] The specific functions and examples of the modules and sub-modules of the system of the embodiments of the present application are described in the above-mentioned method embodiments, and the corresponding descriptions are not repeated here.

[0078] In the technical solution of the present application, the acquisition, storage and application of user personal information comply with relevant laws and regulations and do not violate public order and good customs.

[0079] According to the embodiments of the present application, a system and a readable storage medium are also provided.

[0080] Figure 6 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application described and / or claimed in this document.

[0081] As shown in Figure 6 The electronic device 800 includes a computing unit 801 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. Various programs and data required for the operation of the electronic device 800 can also be stored in the RAM 803. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other through a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0082] A plurality of components in the electronic device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the electronic device 800 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0083] The computing unit 801 can be various general and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 801 performs various methods and processes described above, such as the image processing method applicable to tunnel detection. For example, in some embodiments, the image processing method applicable to tunnel detection can be implemented as a computer software program, which is tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded onto the RAM 803 and executed by the computing unit 801, one or more steps of the image processing method applicable to tunnel detection described above can be performed. Alternatively, in other embodiments, the computing unit 801 can be configured to perform the image processing method applicable to tunnel detection by any other appropriate means, such as by means of firmware.

[0084] The various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0085] Program code for carrying out operations of the methods of the present application can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, and partially on a remote machine or entirely on a remote machine or server.

[0086] In the context of the present application, a machine-readable medium can be a tangible medium that can contain or store program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable storage medium can include, without limitation, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the foregoing.

[0087] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0088] The systems and techniques described herein can be implemented in a computing system that includes a back end component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a front end component, e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described herein, or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0089] The computer system can include clients and servers. The clients and the servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0090] It should be understood that the steps shown in the various forms above can be reordered, added to, or removed. For example, the steps described in the present disclosure can be performed in parallel, in series, or in a different order, without limitation herein, as long as the desired results of the technology disclosed in the present disclosure are achieved.

[0091] The specific embodiments described above are not intended to limit the scope of the present application. It will be apparent to those of ordinary skill in the art that various modifications, combinations, sub-combinations and alternatives can be made to the specific embodiments without departing from the principles of the present application. Any modifications, equivalent substitutions, improvements, and the like which are within the principles of the present application are intended to be included in the scope of the present application.

Claims

1. An image processing method suitable for the detection of a hydraulic tunnel, characterized in that, The method comprises: determining a vault region curvature feature based on a vault region image in a first tunnel section image and a vault region image in a second tunnel section image in two adjacent tunnel section images, wherein there is a field of view overlap region between the two adjacent tunnel section images; determining an irregular settlement region in a tunnel surface region corresponding to the two adjacent tunnel section images and a curvature feature of the irregular settlement region based on the vault region curvature feature; adjusting the field of view overlap region between the two adjacent tunnel section images based on the curvature feature of the irregular settlement region; extracting a first feature point set from the first tunnel section image and a second feature point set from the second tunnel section image based on the adjusted field of view overlap region; in a case where a matching degree between the first feature point set and the second feature point set meets a preset first matching condition, performing image stitching on the first tunnel section image and the second tunnel section image based on the first feature point set and the second feature point set to obtain a tunnel stitching image.

2. The method of claim 1, wherein, The method comprises: extracting a plurality of first vault profile line segments from the vault region image in the first tunnel section image, and performing circular arc curve fitting on each first vault profile line segment to obtain each first circular arc curve; determining a first vault region curvature distribution map based on a curvature corresponding to each first circular arc curve and a position of each first circular arc curve in the vault region image; extracting a plurality of second vault profile line segments from the vault region image in the second tunnel section image, and performing circular arc curve fitting on each second vault profile line segment to obtain each second circular arc curve; determining a second vault region curvature distribution map based on a curvature corresponding to each second circular arc curve and a position of each second circular arc curve in the vault region image; determining the vault region curvature feature based on the first vault region curvature distribution map and the second vault region curvature distribution map.

3. The method of claim 2, wherein, The method comprises: determining a third vault region curvature distribution map based on the vault region curvature feature; determining a vault region curvature change rate and a curvature change direction in each continuous sampling interval in the third vault region curvature distribution map; in a case where the vault region curvature change rate in a first sampling interval is greater than a preset change rate threshold and the curvature change direction meets a preset direction change condition, determining the first sampling interval as an irregular settlement region, and determining a curvature feature of the irregular settlement region based on the vault region curvature change rate and the curvature change direction of the first sampling interval.

4. The method of claim 3, wherein, The adjusting the field-of-view overlap region between the two adjacent tunnel section images based on the curvature feature of the irregular settlement region comprises: determining a curvature variation range of the irregular settlement region based on a curvature variation rate of a vault region in the curvature feature of the irregular settlement region; determining a field-of-view overlap ratio based on the curvature variation range; determining an adjusted position of the field-of-view overlap region based on position information of the irregular settlement region and a boundary position of the field-of-view overlap region between the two adjacent tunnel section images; and adjusting the adjusted position of the field-of-view overlap region based on the field-of-view overlap ratio to obtain an adjusted field-of-view overlap region.

5. The method of claim 1, wherein, The image splicing the first tunnel section image and the second tunnel section image based on the first feature point set and the second feature point set to obtain a tunnel splicing image comprises: performing feature fusion on the first feature point set and the second feature point set to obtain a third feature point set; splicing the first tunnel section image and the second tunnel section image based on the third feature point set to obtain a tunnel splicing image.

6. The method of claim 5, wherein, The feature fusion on the first feature point set and the second feature point set to obtain a third feature point set comprises: in a case where a matching degree of a first feature point in the first feature point set and a second feature point in the second feature point set is greater than a second matching condition, fusing the first feature point and the second feature point to obtain a third feature point, and adding the third feature point to the third feature point set; in a case where a matching degree of a third feature point in the first feature point set and a fourth feature point in the second feature point set is less than the second matching condition, adding the third feature point and the fourth feature point to the third feature point set.

7. The method of claim 1, wherein, Further comprising: in a case where a matching degree between the first feature point set and the second feature point set does not conform to the first matching condition, determining an adjustment mode of the field-of-view overlap region based on the matching degree between the first feature point set and the second feature point set; and re-adjusting the field-of-view overlap region between the two adjacent tunnel section images based on the adjustment mode of the field-of-view overlap region.

8. An image processing device suitable for use in the detection of a water tunnel, characterized in that Comprise: a first feature determination module configured to determine a vault region curvature feature based on a vault region image in a first tunnel section image and a vault region image in a second tunnel section image in two adjacent tunnel section images, wherein there is a field-of-view overlap region between the two adjacent tunnel section images; a second feature determination module configured to determine an irregular settlement region in a tunnel surface region corresponding to the two adjacent tunnel section images and a curvature feature of the irregular settlement region based on the vault region curvature feature; an overlap region adjustment module configured to adjust a field-of-view overlap region between the two adjacent tunnel section images based on the curvature feature of the irregular settlement region. The feature point extraction module is configured to extract a first feature point set from the first tunnel profile image and a second feature point set from the second tunnel profile image based on the adjusted field of view overlap region; The image stitching module is configured to perform image stitching on the first tunnel profile image and the second tunnel profile image based on the first feature point set and the second feature point set when a matching degree between the first feature point set and the second feature point set meets a preset first matching condition, to obtain a tunnel spliced image.

9. An image processing system suitable for tunnel detection, characterized in that, The method comprises: at least one processor, and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the processor is configured to obtain the instructions from the memory and execute the instructions to enable the at least one processor to perform the method of any one of claims 1-7.

10. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are configured to be provided to a computer to instruct the computer to perform the method of any one of claims 1-7.

Images