CN-121999297-A - Drainage pipeline deposition defect judging method based on image illumination difference verification

CN121999297ACN 121999297 ACN121999297 ACN 121999297ACN-121999297-A

Abstract

The invention relates to the technical field of drainage pipeline detection, in particular to a drainage pipeline deposition defect judging method based on image illumination difference verification, which comprises the steps of acquiring an image sequence in a drainage pipeline and establishing an along-path position index; the method comprises the steps of performing deposition defect identification to obtain a candidate deposition area, judging whether the deposition defect exists or not, responding to the existence of the deposition defect, verifying whether the deposition defect is misjudged based on a preset dimming process, determining the grade of the deposition defect according to a standard when the verification is not misjudged, and carrying out self-verification on identification results under complex illumination, water surface and reflective environment without relying on manual review, so that the reliability and engineering usability of the determination result of the deposition defect are enhanced, and the problem that the misjudgment result directly enters a defect grade evaluation link to cause determination deviation is avoided.

Inventors

ZHANG YUANDI
LINGHU JIN
ZHU QINGHAI
ZHANG XIANG
ZHANG TIANQIN
YANG JIE
Yu Huangyan
HAO ZHONGFENG
WANG MAN
ZHENG YUEJUN
LIU CONG
HU YIN
XIE WENMIN
JIN SHAOFEI
XIAO CHENGLONG

Assignees

重庆市测绘科学技术研究院

Dates

Publication Date: 20260508
Application Date: 20260210

Claims (10)

1. The drainage pipeline deposition defect judging method based on image illumination difference verification is characterized by comprising the following steps of: Acquiring an image sequence in a drainage pipeline, and establishing an along-path position index for the image sequence; performing deposition defect identification based on the image sequence to obtain a candidate deposition area, and judging whether the deposition defect exists or not; Responding to existence of a deposition defect, and verifying whether the deposition defect is misjudged based on a preset dimming process; When the fact that misjudgment does not exist is verified, determining the grade of the deposition defect according to a standard; The verification method comprises the steps of obtaining multi-frame illumination difference images according to the preset dimming process at the same position index corresponding to the candidate deposition area, respectively extracting the deposition area from each frame of illumination difference images to form a deposition area change result, and outputting a misjudgment verification result according to the stability of the deposition area change result and the correlation with reflection interference or water surface interference.
2. The method for determining the deposition defect of the drainage pipeline based on the image illumination difference verification according to claim 1, wherein the preset light change process is an illumination state sequence, each illumination state in the illumination state sequence at least comprises an illumination intensity parameter and an illumination incidence direction parameter, and the illumination state sequence at least comprises changes of different illumination intensities and changes of different incidence directions.
3. The method for judging the deposition defect of the drainage pipeline based on the image illumination difference verification according to claim 1, wherein the deposition area change result comprises overlapping consistency among deposition areas of frames, drifting degree of boundaries of the deposition areas, fluctuation of the duty ratio of a highlight reflection area in a candidate deposition area and fluctuation of the relative relation between the boundaries of the deposition areas and a water line.
4. The method for judging the deposition defect of the drainage pipeline based on the image illumination difference verification according to claim 3, wherein the step of outputting the misjudgment verification result comprises the steps of judging that the morphology of the deposition area is unstable and misjudgment exists when the overlapping consistency among the deposition areas of each frame is lower than a preset consistency threshold value or the boundary drift degree of the deposition area is abnormal; when the morphology of the deposition area is stable and at least one of the following is satisfied, determining that there is a false determination: the duty ratio fluctuation of the highlight reflective area in the candidate deposition area is abnormal; The relative relationship between the boundary of the deposition area and the water surface line fluctuates abnormally.
5. The method for judging deposition defects of a drain pipeline based on image illumination difference verification according to claim 4, wherein when misjudgment exists in verification, further recognition processing is triggered to obtain a corrected deposition area and regenerate a deposition defect level, and further recognition results except the deposition defect level are output; The further identification result at least comprises a misjudgment type identification, a corrected deposition area and a confidence representation value of the corrected deposition area, wherein the misjudgment type identification comprises a reflective interference type, a water surface interference type and an unknown interference type, misjudgment situations with the fluctuation abnormality of the duty ratio of the highlight reflective area in the candidate deposition area are identified as reflective interference types, misjudgment situations with the fluctuation abnormality of the relative relation between the boundary of the deposition area and the water surface line are identified as water surface interference types, and the rest misjudgment situations are identified as unknown interference types.
6. The method for judging sedimentation defect of a drainage pipeline based on image illumination difference verification according to claim 5, wherein when the misjudgment type is identified as the reflection interference type, performing highlight reflection inhibition processing on each frame of illumination difference image, re-extracting a sedimentation region based on each frame of image after inhibition processing, performing multi-frame fusion on the re-extracted sedimentation region, and outputting a corrected sedimentation region.
7. The drainage pipeline deposition defect judging method based on image illumination difference verification according to claim 5 is characterized by extracting water surface lines from each frame of illumination difference images and performing position alignment when the misjudgment type is identified as water surface interference type, introducing constraints related to the water surface lines in the process of extracting the deposition areas so as to reduce misdetection caused by long-term fitting of the deposition areas and the water surface lines, fusing the constrained multi-frame deposition areas, and outputting corrected deposition areas.
8. The method for judging sedimentation defects of a drainage pipeline based on image illumination difference verification according to claim 5, wherein when the misjudgment type is identified as an unknown interference type, an image subset with definition reaching standards is selected from the multi-frame illumination difference images, sedimentation areas are extracted from the image subset and fused, and corrected sedimentation areas are output.
9. The method for determining the deposition defect of the drainage pipeline based on the image illumination difference verification according to claim 5, wherein when the confidence representation value is lower than a preset confidence threshold value, adjusting the preset dimming process at the subsequent position index and outputting a rechecking mark, specifically comprises at least one of increasing the illumination intensity span at the subsequent position index, increasing the number of illumination areas in the incident direction and adjusting the exposure parameters.
10. The method for determining the deposition defect of the drainage pipeline based on the image illumination difference verification according to claim 1, wherein the determining the level of the deposition defect according to the standard comprises: Identifying a visible contour line of the extraction pipeline; fitting a circular contour line of the pipeline according to the visible contour line part; Calculating the lengths of two diameter lines in the vertical direction and the horizontal direction according to the circular contour line, and taking an average value as the diameter of the pipeline; calculating the distance between the top of the pipeline and the deposition surface based on the determined deposition area, and further determining the deposition thickness; the duty ratio is calculated by the deposition thickness and the pipe diameter, and the defect level is determined according to the duty ratio.

Description

Drainage pipeline deposition defect judging method based on image illumination difference verification Technical Field The invention relates to the technical field of drainage pipeline detection, in particular to a drainage pipeline deposition defect judging method based on image illumination difference verification. Background Drainage pipe deposition refers to a layer of sediment formed inside the drainage pipe due to long-term accumulation of solid particles, organics, or other impurities carried by the water flow. This phenomenon significantly reduces the water passing capacity of the pipeline, and is one of the main causes of urban inland inundation, sewage overflow and pipeline corrosion. The deposition is explicitly listed as a functional defect in the industry standard of technical specification for urban drainage pipeline detection and evaluation, and is classified into a plurality of grades according to the percentage of the thickness of the deposition to the diameter of the pipeline. For the detection of the deposition, a CCTV (ClosedCircuitTelevision) technology is generally adopted for detection, the technology finishes the shooting task by means of a robot device with a camera entering the pipeline, and the shot image data is transmitted to a remote image processing terminal through a multi-core cable, and the acquired video image data is manually interpreted by a professional person to classify the pipeline defects and finish the assessment of the damage level. However, the manual judgment is greatly influenced by the detected operators, the judgment of the same operator in different environments and different states can be different, and the judgment of different operators is more difficult to unify. And the efficiency of manual judgment is low, the rechecking difficulty is extremely high, and the recognition accuracy is low. On the basis, a detection method based on image recognition is developed, for example, in the prior art, chinese patent publication No. CN113160210A discloses a detection method and a detection device of a drainage pipeline defect based on a depth camera, and the method comprises the following steps of S1, acquiring a first image of the inner surface of the drainage pipeline shot by the depth camera; the method comprises the steps of S2, classifying a first image according to a pre-trained first image classification model, S3, acquiring a second image shot by a depth camera and synchronous with the first image if the first image is classified as a first defect image, S4, classifying the second image according to a pre-trained second image classification model, S5, performing image fusion on the first defect image and the second defect image to generate a third image if the second image is classified as a second defect image, S6, detecting the defect type of the third image according to a pre-trained defect detection model, and outputting a detection result. However, the existing defect image recognition technology similar to the above technology lacks further verification on the aspect of image acquisition, so that image input caused by abnormal acquisition due to the influence of light water flow can occur, and further, the calculation force is wasted and the accuracy is reduced. Disclosure of Invention The invention aims to provide a drainage pipeline deposition defect judging method based on image illumination difference verification, which is used for solving the problems that the existing defect image identification lacks further verification on the image acquisition layer, and further, the acquisition abnormal image input is caused by the influence of light water flow, so that the calculation force is wasted and the accuracy is reduced. Therefore, the invention provides a drainage pipeline deposition defect judging method based on image illumination difference verification, which comprises the following steps: Acquiring an image sequence in a drainage pipeline, and establishing an along-path position index for the image sequence; performing deposition defect identification based on the image sequence to obtain a candidate deposition area, and judging whether the deposition defect exists or not; Responding to existence of a deposition defect, and verifying whether the deposition defect is misjudged based on a preset dimming process; When the fact that misjudgment does not exist is verified, determining the grade of the deposition defect according to a standard; The verification method comprises the steps of obtaining multi-frame illumination difference images according to the preset dimming process at the same position index corresponding to the candidate deposition area, respectively extracting the deposition area from each frame of illumination difference images to form a deposition area change result, and outputting a misjudgment verification result according to the stability of the deposition area change result and the correlation with reflection interference o