CN111127399A - Underwater pier disease identification method based on deep learning and sonar imaging - Google Patents
Underwater pier disease identification method based on deep learning and sonar imaging Download PDFInfo
- Publication number
- CN111127399A CN111127399A CN201911195800.2A CN201911195800A CN111127399A CN 111127399 A CN111127399 A CN 111127399A CN 201911195800 A CN201911195800 A CN 201911195800A CN 111127399 A CN111127399 A CN 111127399A
- Authority
- CN
- China
- Prior art keywords
- disease
- deep learning
- underwater
- sonar
- picture
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 201000010099 disease Diseases 0.000 title claims abstract description 37
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000003384 imaging method Methods 0.000 title claims abstract description 23
- 238000013135 deep learning Methods 0.000 title claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 14
- 238000012360 testing method Methods 0.000 claims abstract description 12
- 238000012795 verification Methods 0.000 claims abstract description 6
- 238000002372 labelling Methods 0.000 claims description 9
- 230000007547 defect Effects 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 6
- 235000002566 Capsicum Nutrition 0.000 claims description 3
- 239000006002 Pepper Substances 0.000 claims description 3
- 235000016761 Piper aduncum Nutrition 0.000 claims description 3
- 235000017804 Piper guineense Nutrition 0.000 claims description 3
- 235000008184 Piper nigrum Nutrition 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000009432 framing Methods 0.000 claims description 3
- 150000003839 salts Chemical class 0.000 claims description 3
- 244000203593 Piper nigrum Species 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 abstract description 5
- 230000009189 diving Effects 0.000 abstract description 3
- 238000012216 screening Methods 0.000 abstract description 2
- 241000722363 Piper Species 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000009991 scouring Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 210000002435 tendon Anatomy 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
- Image Processing (AREA)
Abstract
The invention provides an underwater pier disease identification method based on deep learning and sonar imaging, which comprises the steps of obtaining and preparing underwater pier scanning pictures including disease and disease-free pictures by using an underwater side sonar device; increasing the number of data sets by using an image enhancement method; marking the data set, marking the disease area by using a rectangular frame and storing coordinate information; dividing a data set into a training test set, a verification set and a test set; establishing a yolov3 model in a deep learning target detection network, and training to obtain a training model; the control side sonar equipment on the surface of water scans along pier part under water, obtains the scanning picture, utilizes the good yolov3 model of training to carry out pier disease automatic identification under water. The invention has high efficiency and low cost, and has obvious advantages of automation and real-time performance compared with the traditional manual diving method and sonar manual screening method.
Description
Technical Field
The invention relates to the technical field of civil engineering and artificial intelligence interaction, in particular to an underwater pier disease identification method based on deep learning and sonar imaging.
Background
The bridge pier is a main bearing component of the bridge pier, and most of load of the bridge structure is transferred to the foundation through the bridge pier. Since the loss of the bearing capacity of any pier leads to the overall instability and destruction of the pier, the safety of the pier must be highly regarded. The underwater part of the bridge pier is subjected to the action of severe environments such as scouring and corrosion for a long time, so that defects such as defects, cracks and exposed ribs can be generated underwater and even damaged, and the service life and even the bearing capacity of the bridge can be seriously influenced. At present, a detection method of an underwater pier part is also an artificial diving method and a sonar equipment scanning method, the artificial detection method is time-consuming, labor-consuming, high in manufacturing cost and low in automation degree, the sonar equipment scanning method generates a large number of pictures, manual identification efficiency is low, and misjudgment are easy, so that an automatic identification method of an underwater pier disease is urgently needed.
Deep learning is one of machine learning methods, and is applied to the visual field, image classification, target detection, semantic segmentation and the like. The sonar equipment presents the condition of the underwater pier in an acoustic imaging mode, and can overcome the defect that optical imaging is influenced by illumination and water turbidity degree, the underwater pier disease real-time identification method based on the deep learning yolov3 network and sonar imaging provided by the invention carries out deep learning training on a sonar imaging picture by combining the advantages of deep learning and sonar imaging, and can automatically identify diseases and identify the positions of the diseases by using a target detection model, so that the efficiency and the cost of underwater pier detection are greatly improved, the model can reach the identification speed of 30FPS, and the requirement of real-time detection can be met.
Disclosure of Invention
The purpose of the invention is as follows: the invention provides an underwater pier disease identification method based on deep learning and sonar imaging, which is high in efficiency and low in cost, and has obvious advantages of automation and real-time performance compared with a traditional manual diving method and a sonar manual screening method.
The technical scheme is as follows: the invention relates to a method for enhancing the resolution of an underwater pier disease image based on deep learning, which comprises the following steps:
(1) acquiring an underwater pier disease picture and a picture in a normal state by using side sonar equipment to form a data set;
(2) expanding the data set by using a data enhancement method, and labeling each picture;
(3) dividing a data set into a training test set, a verification set and a test set;
(4) establishing a deep learning yolov3 model, training the underwater pier sonar imaging data set in the step (1), and storing the trained yolov3 model and parameters;
(5) on site, scanning an underwater pier by using sonar equipment, and removing salt and pepper noise of an image by using an image mean filtering method and a Gaussian filtering method;
(6) and (4) automatically identifying whether the picture obtained in the step (5) is a disease picture by using the convolutional neural network model which is obtained in the step (4) and can automatically identify the disease of the sonar imaging picture of the underwater bridge pier, and framing the position of the disease.
Further, the disease picture in the step (1) mainly comprises a defect picture, a crack picture and a bare picture.
Further, the picture size in step (1) is 1200x1200 pixels.
Further, the labeling in the step (2) adopts a rectangular frame.
Further, the ratio of the training test set, the verification set and the test set in the step (3) is 8:1: 1.
Further, the yolov3 model in the step (4) adopts a VGG network model, the network structure adopts yolov3380 and yolov3512 models, and the network input is 380 or 512 pixel size.
Has the advantages that: compared with the prior art, the invention has the beneficial effects that: the image recognition method is high in efficiency, low in cost, high in automation degree, capable of recognizing in real time, more concise and effective compared with the traditional manual detection method, and capable of greatly improving the accuracy and the automation degree of image recognition compared with the existing automation methods such as underwater robots, sonar imaging and the like.
Drawings
FIG. 1 is a flow chart of the present invention;
fig. 2 is a schematic diagram of yolov3 network structure used in the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings. As shown in fig. 1, an underwater pier disease identification method based on deep learning and sonar imaging comprises the following steps:
1. acquiring an underwater pier disease picture and a picture in a normal state by using side sonar equipment to form a data set, and setting each picture to be 1200x1200pixel size; the disease pictures mainly comprise defect pictures, crack pictures and exposed tendon pictures.
2. Expanding the data set by using a data enhancement method, and labeling each picture; the data enhancement method adopts random rotation and random cutting, the labeling adopts rectangular frame labeling, and the form of the labeling frame is recorded in the form of an xml file.
3. The data set is divided into a training test set, a verification set and a test set according to the ratio of 8:1: 1.
4. Establishing a deep learning yolov3 model, as shown in fig. 2, training an underwater pier sonar imaging data set in the step S1, and storing the trained yolov3 model and parameters; the yolov3 network adopts a VGG network model, and the network structures respectively adopt yolov3380 and yolov3512 models.
5. On site, sonar equipment is used for scanning the underwater bridge pier, and the salt and pepper noise of the image is removed by using an image mean filtering method and a Gaussian filtering method.
6. And (4) automatically identifying whether the picture obtained in the step (5) is a disease picture or not by using the convolutional neural network model which is obtained in the step (4) and can automatically identify the disease of the sonar imaging picture of the underwater bridge pier, and framing the position of the disease.
Laboratory conditions required for training: GTX 1060 video card, Windows system, Python programming language, Tensorflow deep learning framework.
Data set: mainly contain the image data set of pier sonar under water formation of image, including disease pictures such as defect, crack, exposure muscle and the picture under the normal condition.
The above description is only of the preferred embodiments of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.
Claims (6)
1. The underwater pier disease identification method based on deep learning and sonar imaging is characterized by comprising the following steps of:
(1) acquiring an underwater pier disease picture and a picture in a normal state by using side sonar equipment to form a data set;
(2) expanding the data set by using a data enhancement method, and labeling each picture;
(3) dividing a data set into a training test set, a verification set and a test set;
(4) establishing a deep learning yolov3 model, training the underwater pier sonar imaging data set in the step (1), and storing the trained yolov3 model and parameters;
(5) on site, scanning an underwater pier by using sonar equipment, and removing salt and pepper noise of an image by using an image mean filtering method and a Gaussian filtering method;
(6) and (4) automatically identifying whether the picture obtained in the step (5) is a disease picture by using the convolutional neural network model which is obtained in the step (4) and can automatically identify the disease of the sonar imaging picture of the underwater bridge pier, and framing the position of the disease.
2. The underwater pier disease identification method based on deep learning and sonar imaging according to claim 1, wherein the disease pictures in the step (1) mainly include defect pictures, crack pictures and exposed rib pictures.
3. The underwater pier disease identification method based on deep learning and sonar imaging according to claim 1, wherein the size of the picture in step (1) is 1200x1200 pixels.
4. The underwater pier disease identification method based on deep learning and sonar imaging according to claim 1, wherein the labeling in step (2) is performed by rectangular frame labeling.
5. The underwater pier disease identification method based on deep learning and sonar imaging according to claim 1, wherein the proportion of the training test set, the verification set and the test set in the step (3) is 8:1: 1.
6. The underwater pier disease identification method based on deep learning and sonar imaging according to claim 1, wherein the yolov3 model in the step (4) is a VGG network model, the network structure is a yolov3380 model and a yolov3512 model, and the network input is 380 or 512 pixels in size.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911195800.2A CN111127399A (en) | 2019-11-28 | 2019-11-28 | Underwater pier disease identification method based on deep learning and sonar imaging |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911195800.2A CN111127399A (en) | 2019-11-28 | 2019-11-28 | Underwater pier disease identification method based on deep learning and sonar imaging |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111127399A true CN111127399A (en) | 2020-05-08 |
Family
ID=70497048
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911195800.2A Pending CN111127399A (en) | 2019-11-28 | 2019-11-28 | Underwater pier disease identification method based on deep learning and sonar imaging |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111127399A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111681240A (en) * | 2020-07-07 | 2020-09-18 | 福州大学 | Bridge surface crack detection method based on YOLO v3 and attention mechanism |
| CN111832607A (en) * | 2020-05-28 | 2020-10-27 | 东南大学 | Bridge disease real-time detection method based on model pruning |
| CN112285682A (en) * | 2020-10-20 | 2021-01-29 | 水利部交通运输部国家能源局南京水利科学研究院 | 360-degree multi-beam sonar scanning device and method for hydraulic engineering culvert environment |
| CN112508901A (en) * | 2020-12-01 | 2021-03-16 | 广州大学 | Underwater structure disease identification method, system and device and storage medium |
| CN113313107A (en) * | 2021-04-25 | 2021-08-27 | 湖南桥康智能科技有限公司 | Intelligent detection and identification method for multiple types of diseases on cable surface of cable-stayed bridge |
| CN113450343A (en) * | 2021-07-19 | 2021-09-28 | 福州大学 | Sonar imaging based depth learning and intelligent detection method for crack diseases of planar pile pier |
| CN113566753A (en) * | 2021-07-19 | 2021-10-29 | 福州大学 | Measuring point layout method based on mechanical scanning imaging sonar scanning bridge pier foundation scouring |
| CN114494261A (en) * | 2022-04-18 | 2022-05-13 | 陕西易合交通科技有限公司 | Underwater structure disease data processing method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544533A (en) * | 2018-11-23 | 2019-03-29 | 聚时科技(上海)有限公司 | A kind of metal plate defect detection and measure based on deep learning |
| CN110189317A (en) * | 2019-05-30 | 2019-08-30 | 上海卡罗网络科技有限公司 | A kind of road image intelligent acquisition and recognition methods based on deep learning |
| CN110348342A (en) * | 2019-06-27 | 2019-10-18 | 广东技术师范大学天河学院 | A kind of piping disease image partition method based on full convolutional network |
-
2019
- 2019-11-28 CN CN201911195800.2A patent/CN111127399A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544533A (en) * | 2018-11-23 | 2019-03-29 | 聚时科技(上海)有限公司 | A kind of metal plate defect detection and measure based on deep learning |
| CN110189317A (en) * | 2019-05-30 | 2019-08-30 | 上海卡罗网络科技有限公司 | A kind of road image intelligent acquisition and recognition methods based on deep learning |
| CN110348342A (en) * | 2019-06-27 | 2019-10-18 | 广东技术师范大学天河学院 | A kind of piping disease image partition method based on full convolutional network |
Non-Patent Citations (3)
| Title |
|---|
| JOSEPH REDMON ALI FARHADI: "YOLOv3: An Incremental Improvement", 《ARXIV:1804.02767V1 [CS.CV]》 * |
| 译夫: "yolov3自学笔记（一）", 《HTTPS://ZHUANLAN.ZHIHU.COM/P/77061164》 * |
| 郭龙源等: "基于 YOLOv3 的磁瓦表面缺陷检测算法", 《成都工业学院学报》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111832607A (en) * | 2020-05-28 | 2020-10-27 | 东南大学 | Bridge disease real-time detection method based on model pruning |
| CN111681240A (en) * | 2020-07-07 | 2020-09-18 | 福州大学 | Bridge surface crack detection method based on YOLO v3 and attention mechanism |
| CN112285682A (en) * | 2020-10-20 | 2021-01-29 | 水利部交通运输部国家能源局南京水利科学研究院 | 360-degree multi-beam sonar scanning device and method for hydraulic engineering culvert environment |
| CN112508901A (en) * | 2020-12-01 | 2021-03-16 | 广州大学 | Underwater structure disease identification method, system and device and storage medium |
| CN112508901B (en) * | 2020-12-01 | 2024-04-05 | 广州大学 | Underwater structure disease identification method, system, device and storage medium |
| CN113313107A (en) * | 2021-04-25 | 2021-08-27 | 湖南桥康智能科技有限公司 | Intelligent detection and identification method for multiple types of diseases on cable surface of cable-stayed bridge |
| CN113450343A (en) * | 2021-07-19 | 2021-09-28 | 福州大学 | Sonar imaging based depth learning and intelligent detection method for crack diseases of planar pile pier |
| CN113566753A (en) * | 2021-07-19 | 2021-10-29 | 福州大学 | Measuring point layout method based on mechanical scanning imaging sonar scanning bridge pier foundation scouring |
| CN114494261A (en) * | 2022-04-18 | 2022-05-13 | 陕西易合交通科技有限公司 | Underwater structure disease data processing method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111127399A (en) | Underwater pier disease identification method based on deep learning and sonar imaging | |
| Wang et al. | Automatic detection of wind turbine blade surface cracks based on UAV-taken images | |
| EP3739513A1 (en) | Surface defect identification method and apparatus | |
| CN113971779B (en) | Water gauge automatic reading method based on deep learning | |
| CN112598672A (en) | Pavement disease image segmentation method and system based on deep learning | |
| CN110853041A (en) | Underwater pier component segmentation method based on deep learning and sonar imaging | |
| DE112017007778T5 (en) | Method and device for detecting a pipeline and storage medium | |
| CN110751630B (en) | Power transmission line foreign matter detection method and device based on deep learning and medium | |
| CN111127448B (en) | Method for detecting air spring fault based on isolated forest | |
| CN111060521B (en) | Method for detecting quantity and area of air bubbles on surface of precast concrete T beam | |
| JP2018185552A (en) | Image analysis apparatus, image analysis method, and program | |
| CN110288623B (en) | Data compression method for unmanned aerial vehicle maritime net cage culture inspection image | |
| CN109978847B (en) | Automatic inhaul cable sleeve disease identification method based on transfer learning and inhaul cable robot | |
| CN113155860A (en) | Flow state video monitoring-based method and system for diagnosing structural damage of water-passing building | |
| CN111079724B (en) | Unmanned aerial vehicle-based sea floating garbage identification method | |
| CN112101138B (en) | Bridge inhaul cable surface defect real-time identification system and method based on deep learning | |
| CN112727705A (en) | Monitoring and flaw detection method for blades of wind generating set | |
| CN110826391A (en) | Bleeding area detection method, bleeding area detection system, computer device and storage medium | |
| CN109521021A (en) | A kind of nuclear power plant containment shell appearance inspecting system and method | |
| CN116484259A (en) | Urban pipe network defect position positioning analysis method and system | |
| CN111275020A (en) | Room state identification method | |
| JPH09119900A (en) | Method for automatically evaluating concrete defect | |
| CN107330440A (en) | Sea state computational methods based on image recognition | |
| CN112669269A (en) | Pipeline defect classification and classification method and system based on image recognition | |
| Vijayan et al. | A survey on surface crack detection in concretes using traditional, image processing, machine learning, and deep learning techniques |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200508 |
|
| RJ01 | Rejection of invention patent application after publication |