CN-120431029-B - Concrete defect detection method, detection device, electronic equipment and storage medium

CN120431029BCN 120431029 BCN120431029 BCN 120431029BCN-120431029-B

Abstract

The application discloses a concrete defect detection method, a detection device, electronic equipment and a storage medium, wherein the concrete defect detection method comprises the steps of responding to a received detection instruction, acquiring a current concrete image of concrete, wherein the detection instruction is used for indicating to detect the defect of the concrete; inputting the current concrete image to a target defect detection model to obtain a corresponding defect detection result, wherein the target defect detection model is obtained based on a single-step multi-frame target detection model and a texture enhancement model, and the texture enhancement model is used for enhancing and extracting texture characteristics of the current concrete image. The method realizes the defect detection of the concrete based on the target defect detection model constructed by the single-step multi-frame target detection model and the texture enhancement model, has short detection time, and improves the detection efficiency and detection accuracy of the defect detection of the concrete.

Inventors

LUO QILING
LI MINGCHAO
ZHANG MENGXI
ZHANG JINRUI
MIN QIAOLING
TIAN DAN

Assignees

天津大学

Dates

Publication Date: 20260512
Application Date: 20250410

Claims (8)

1. The concrete defect detection method is characterized by being applied to a concrete defect detection system, wherein the concrete defect detection system comprises a camera and processing equipment: The processing equipment acquires an initial defect detection model, wherein the initial defect detection model is obtained by splicing a texture enhancement model between a visual geometry group module of a single-step multi-frame target detection model and a top-down module in a characteristic pyramid network, the texture enhancement model is used for enhancing and extracting texture features of a concrete image, the texture enhancement model is obtained by fusing a central difference convolution module, a horizontal difference convolution module, a vertical difference convolution module, an angle difference convolution module and a channel shuffling module, the central difference convolution module, the horizontal difference convolution module, the vertical difference convolution module, the angle difference convolution module and the channel shuffling module are sequentially connected, the central difference convolution module, the horizontal difference convolution module and the vertical difference convolution module are connected with the channel shuffling module through residual errors, the central difference between a central pixel of a current concrete image and a surrounding neighborhood is calculated, the central pixel is used for representing a pixel point of the central position of the current concrete image, the horizontal difference convolution module is used for extracting texture features of a first texture feature map, the horizontal difference convolution module is used for extracting texture features of a third texture feature map, the horizontal difference convolution module is used for obtaining texture feature map of a third texture feature map, the third texture feature map is obtained by a fourth texture feature map, the third texture feature map is obtained by a third channel feature map, the third texture feature map is obtained by the third channel feature map is obtained by the vertical difference convolution module, and the third texture feature map is used for obtaining texture feature map of the third texture feature map, obtaining a fifth characteristic diagram; The processing equipment trains the initial defect detection model by using a training set to obtain a target defect detection model, wherein the training set comprises at least one historical concrete defect image, and the historical concrete defect image is provided with marks which mainly mark defect types, defect positions and defect sizes in the historical concrete defect image; The processing equipment responds to the received detection instruction and sends a first acquisition instruction to the camera, wherein the first acquisition instruction is used for triggering the camera to acquire a current concrete image of concrete, the camera responds to the first acquisition instruction and acquires and sends the current concrete image of the concrete to the processing equipment, and the detection instruction is used for indicating the defect detection of the concrete; The processing equipment inputs the current concrete image to the target defect detection model to obtain a corresponding defect detection result, wherein the defect detection result comprises a first detection result used for representing that the current concrete image contains current defect information and a second detection result used for representing that the current concrete image does not contain the current defect information, the current defect information comprises a defect type, a defect position and a defect size, and when the defect detection result is the first detection result used for representing that the current concrete image contains the current defect information, a defect processing scheme corresponding to the first detection result is obtained.
2. The concrete defect detection method according to claim 1 is characterized in that the center difference convolution module is used for calculating texture feature differences between a center pixel of the current concrete image and surrounding neighborhood to obtain a first feature image, the horizontal difference convolution module is used for extracting texture features of the first feature image in the horizontal direction to obtain a second feature image, the vertical difference convolution module is used for extracting texture features of the second feature image in the vertical direction to obtain a third feature image, the angle difference convolution module is used for extracting texture features of the third feature image in multiple angles to obtain a fourth feature image, and the channel shuffling module is used for enhancing cross-channel communication of the texture features of the first feature image, the second feature image, the third feature image and the fourth feature image to obtain a fifth feature image.
3. The method for detecting a concrete defect according to claim 1, wherein before the training set is input to the initial defect detection model for training to obtain the target defect detection model, the method further comprises: performing data enhancement processing on the training set to obtain an enhanced training set; The training set is input to the initial defect detection model for training, and the target defect detection model is obtained, which comprises the following steps: and inputting the enhanced training set to the initial defect detection model for training to obtain the target defect detection model.
4. The method for detecting a concrete defect according to claim 1, further comprising: When the defect detection result is a first detection result used for representing that the current concrete image contains current defect information, inputting the first detection result into a residual life prediction model to obtain the residual life of the concrete, and training a deep learning neural network model based on historical defect information by the residual life prediction model; and when the residual life of the concrete is smaller than or equal to the residual life threshold value, generating corresponding warning information.
5. The method according to any one of claims 1 to 4, wherein after the obtaining of the defect treatment plan corresponding to the first detection result, the method further comprises: And sending the defect processing scheme to a client associated with the inspector.
6. The concrete defect detection device is characterized by being applied to a concrete defect detection system, wherein the concrete defect detection system comprises a camera and processing equipment, and the processing equipment comprises: The system comprises a fusion unit, a central differential convolution module, a horizontal differential convolution module, a vertical differential convolution module, an angle differential convolution module and a channel shuffling module, wherein the central differential convolution module, the horizontal differential convolution module, the vertical differential convolution module, the angle differential convolution module and the channel shuffling module are sequentially connected, the central differential convolution module, the horizontal differential convolution module and the vertical differential convolution module are connected with the channel shuffling module through residual errors, the texture enhancement module is used for enhancing and extracting texture characteristics of a concrete image, the central differential convolution module is used for calculating texture characteristic difference values between central pixels and surrounding neighborhoods of the current concrete image to obtain a first characteristic image, the central pixels are used for representing pixel points of the central positions of the current concrete image, the horizontal differential convolution module is used for extracting texture characteristics of the first characteristic image in the horizontal direction, and obtaining a second characteristic image; the fusion module is used for splicing the texture enhancement model between a visual geometry group module of the single-step multi-frame target detection model and a top-down module in the characteristic pyramid network to obtain an initial defect detection model; the training module is used for training the initial defect detection model by using a training set to obtain a target defect detection model, wherein the training set comprises at least one historical concrete defect image, and the historical concrete defect image is provided with marks which mainly mark defect types, defect positions and defect sizes in the historical concrete defect image; The first acquisition module is used for responding to the received detection instruction and sending a first acquisition instruction to the camera, wherein the first acquisition instruction is used for triggering the camera to acquire a current concrete image of the concrete, the camera responds to the first acquisition instruction and acquires and sends the current concrete image of the concrete to the processing equipment, and the detection instruction is used for indicating to detect the defect of the concrete; The first input module is used for inputting the current concrete image to the target defect detection model to obtain a corresponding defect detection result, wherein the defect detection result comprises a first detection result used for representing that the current concrete image contains current defect information and a second detection result used for representing that the current concrete image does not contain the current defect information, the current defect information comprises a defect type, a defect position and a defect size, and when the defect detection result is the first detection result used for representing that the current concrete image contains the current defect information, a defect processing scheme corresponding to the first detection result is obtained.
7. An electronic device, comprising: a memory; One or more processors coupled with the memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications configured to perform the concrete defect detection method of any one of claims 1 to 5.
8. A computer readable storage medium having stored therein program code which is callable by a processor to perform the concrete defect detection method according to any one of claims 1 to 5.

Description

Concrete defect detection method, detection device, electronic equipment and storage medium Technical Field The application belongs to the technical field of concrete detection, and particularly relates to a concrete defect detection method, a detection device, electronic equipment and a storage medium. Background The concrete material has good integrity, moldability, durability and fire resistance, and is widely applied to the construction of infrastructures such as house buildings, bridges, tunnels, dams and the like. However, under the coupling effect of factors such as ageing of material performance, external temperature change, long-term load, the concrete is easy to generate defects such as cracks, damages and the like in the construction, operation and use processes, so that the bearing capacity of the concrete is reduced, and the safety risk of the concrete is increased. Therefore, it is important to detect defects in concrete. At present, the defect detection of concrete mainly relies on manual inspection to observe the concrete, and the manual inspection takes a long time, and the manual inspection may have missed detection or misjudgment of the defect. Disclosure of Invention In view of the above, embodiments of the present application provide a method, an apparatus, an electronic device and a storage medium for detecting a concrete defect, so as to overcome or at least partially solve the above problems of the prior art. In a first aspect, an embodiment of the application provides a concrete defect detection method, which comprises the steps of responding to a received detection instruction, obtaining a current concrete image of concrete, wherein the detection instruction is used for indicating to detect defects of the concrete, inputting the current concrete image into a target defect detection model to obtain a corresponding defect detection result, and obtaining the target defect detection model based on a single-step multi-frame target detection model and a texture enhancement model, wherein the texture enhancement model is used for enhancing and extracting texture characteristics of the current concrete image. In a second aspect, an embodiment of the present application provides a concrete defect detection apparatus, where the concrete defect detection apparatus includes a first acquisition module and a first input module. The system comprises a first acquisition module, a first input module and a second input module, wherein the first acquisition module is used for responding to a received detection instruction to acquire a current concrete image of concrete, the detection instruction is used for indicating to detect defects of the concrete, the first input module is used for inputting the current concrete image into a target defect detection model to obtain a corresponding defect detection result, the target defect detection model is obtained based on a single-step multi-frame target detection model and a texture enhancement model, and the texture enhancement model is used for enhancing and extracting texture characteristics of the current concrete image. In a third aspect, an embodiment of the present application provides an electronic device, including a memory, one or more processors coupled to the memory, and one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications configured to perform the method for detecting a concrete defect as provided in the first aspect above. In a fourth aspect, an embodiment of the present application provides a computer readable storage medium having program code stored therein, the program code being callable by a processor to perform the method for detecting a concrete defect as provided in the first aspect above. In a fifth aspect, embodiments of the present application provide a computer program product, which when run on a computer device causes the computer device to perform the method for detecting a concrete defect as provided in the first aspect above. According to the scheme provided by the application, the current concrete image of the concrete is obtained in response to the received detection instruction, the detection instruction is used for indicating the defect detection of the concrete, the current concrete image is input to the target defect detection model, the corresponding defect detection result is obtained, the target defect detection model is obtained based on the single-step multi-frame target detection model and the texture enhancement model, the texture enhancement model is used for enhancing and extracting the texture characteristics of the current concrete image, the defect detection of the concrete based on the target defect detection model constructed by the single-step multi-frame target detection model and the texture enhancement model is realized, the detection time is short, and the detection efficiency an