CN113657151A - Water traffic violation detection method based on YOLO target detection algorithm - Google Patents
Water traffic violation detection method based on YOLO target detection algorithm Download PDFInfo
- Publication number
- CN113657151A CN113657151A CN202110769368.4A CN202110769368A CN113657151A CN 113657151 A CN113657151 A CN 113657151A CN 202110769368 A CN202110769368 A CN 202110769368A CN 113657151 A CN113657151 A CN 113657151A
- Authority
- CN
- China
- Prior art keywords
- ship
- target
- traffic violation
- target detection
- yolo
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 63
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 32
- 230000006399 behavior Effects 0.000 claims abstract description 35
- 238000012544 monitoring process Methods 0.000 claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 2
- 238000000605 extraction Methods 0.000 claims 1
- 238000011897 real-time detection Methods 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention discloses a method for detecting water traffic violation behaviors based on a YOLO target detection algorithm, which comprises the following steps: the method comprises the steps that frames are extracted from a channel monitoring video shot by a monitoring camera, a ship is marked, and a ship target identification data set is manufactured; making a violation behavior data set by cutting a ship image and marking a detection target of traffic violation behaviors, such as a flag, a ship name, a life jacket and the like; and training network parameters by using the data set to obtain a network model for detecting the ship and the illegal behaviors, inputting the monitoring video into the trained network model for detecting the ship and the illegal behaviors, and judging whether the target ship has the traffic illegal behaviors. The invention can realize automatic, accurate and real-time detection of the water traffic violation behaviors through the YOLO target detection algorithm.
Description
Technical Field
The invention relates to the field of pattern recognition, in particular to a method for detecting water traffic violation behaviors based on a YOLO target detection algorithm.
Background
The traditional method for carrying out the reconnaissance on the ship violation behaviors usually needs a large amount of manpower to pay attention to and distinguish, and wastes time and labor. With the rapid development of artificial intelligence, a target detection method can be used, automatic ship violation behavior detection is carried out through a monitoring camera, and the detection efficiency is greatly improved.
At present, target detection technologies for ships mostly pay attention to detection of the ships and pay less attention to violation behaviors. And for the problems of ship name shielding, flag missing and the like in ship violation, the target is too small, so that the occupied pixels in the video are too few, and the detection difficulty is high.
Therefore, the detection of the ship violation behaviors of the small targets can be well realized by using the YOLO target detection algorithm with the added attention mechanism.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide a method for detecting water traffic violation behaviors based on a YOLO target detection algorithm, which can realize real-time automatic detection of small-target ship violation behaviors, thereby improving the efficiency and saving the manpower.
The technical scheme is as follows: in order to realize the purpose of the invention, the technical scheme adopted by the invention is as follows: the invention designs a method for detecting water traffic violation behaviors based on a YOLO target detection algorithm, which comprises the following steps:
step 1: and (4) extracting frames of the channel ship video shot by the monitoring camera, and marking the ship, the ship name and the flag in the channel ship video to be used as a data set.
Step 2: and establishing a YOLO network structure.
And step 3: and training a YOLO network by using the established data set, and adjusting network parameters so as to obtain a ship target detection model and a water traffic violation behavior detection model.
And 4, step 4: inputting the monitoring video into a ship target detection model to detect the position of a target ship in a channel
And 5: and inputting the image of the target ship area into the water traffic violation behavior detection model to obtain a violation behavior detection result of the target ship.
As a further improvement of the present invention, the step (1) specifically comprises: sampling, taking frames and marking the shot channel monitoring video, and establishing a ship target detection data set and a water traffic violation behavior data set.
As a further improvement of the present invention, the step (2) specifically comprises: and constructing a YOLO network which comprises a target detection network and a double attention channel, and obtaining parameters of the network through later training so that the network can detect the names and flags of ships and boats in the video.
As a further improvement of the present invention, the step (3) is specifically: and training the constructed neural network by using the ship target detection data set and the water traffic violation behavior data set to obtain parameters of the network structure.
As a further improvement of the present invention, the step (4) specifically comprises: and inputting the channel monitoring video to be detected into a ship target detection model to obtain a local image of the detected target ship.
As a further improvement of the present invention, the step (5) is specifically: and inputting the detected regional image of the target ship into the water traffic violation detection network model, and outputting the detection result of the violation.
Compared with the prior art, the technical scheme of the invention has the following beneficial technical effects:
(1) the ship violation behaviors can be automatically detected by using a target detection technology, so that the labor is saved, and the efficiency is improved;
(2) the violation behaviors of small targets, such as ship name shielding and flag missing, can be better detected.
(3) Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
Fig. 1 is a schematic flow chart of a water traffic violation detection method based on a YOLO target detection algorithm.
FIG. 2 is a final display result diagram in the embodiment of the present invention.
Detailed Description
The following detailed description of the embodiments of the present invention will be provided with reference to the drawings and examples, so that how to apply the technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented. It should be noted that, as long as there is no conflict, the embodiments and the features of the embodiments of the present invention may be combined with each other, and the technical solutions formed are within the scope of the present invention.
Additionally, the steps illustrated in the flow charts of the figures may be performed in a computer system such as a set of computer-executable instructions and, although a logical order is illustrated in the flow charts, in some cases, the steps illustrated or described may be performed in an order different than here.
The invention designs a method for detecting water traffic violation behaviors based on a YOLO target detection algorithm, which comprises the following steps:
step 1: and (4) extracting frames of the channel ship video shot by the monitoring camera, and marking the ship, the ship name and the flag in the channel ship video to be used as a data set.
Step 2: and establishing a YOLO network structure.
And step 3: and training a YOLO network by using the established data set, and adjusting network parameters so as to obtain a ship target detection model and a water traffic violation behavior detection model. The loss function used in training mainly includes the target positioning offset loss lambda3Lloc(L, g), loss of target confidence Lconf(o, c) and target classification loss Lcla(O, C), the overall loss function is as follows:
L(O,o,C,c,l,g)=λ1Lconf(o,c)+λ2Lcla(O,C)+λ3Lloc(l,g)
wherein λ1,λ2,λ3Is the equilibrium coefficient. Target confidence loss adopts binary cross entropy lossThe loss function is as follows:
wherein o isiIs a binary parameter which represents whether a target really exists in a predicted target boundary box or not,is the sigmoid probability that the target really exists. The objective class loss function also uses binary cross entropy loss, whose loss function is as follows:
wherein O isijIndicating whether the j-type target really exists in the ith prediction target bounding box,is the sigmoid probability that the target really exists. And the target positioning loss function adopts the square sum of the difference between the real deviation value and the predicted deviation value, l is the coordinate offset of the predicted rectangular frame, and g represents the coordinate offset between the group route rectangular frame and the default frame.
And 4, step 4: inputting the monitoring video into a ship target detection model to detect the position of a target ship in a channel
And 5: and inputting the image of the target ship area into the water traffic violation behavior detection model to obtain a violation behavior detection result of the target ship.
Examples
Fig. 1 is a method for detecting water traffic violation based on a YOLO target detection algorithm, which is implemented by the invention, and each step is described in detail below with reference to fig. 1.
In step S110, as the data set of the channel ship and the water traffic violation behaviors is not disclosed, the channel monitoring video shot by self is used, and the channel monitoring video is sampled, framed and labeled, so that a ship target detection data set and a water traffic violation behavior data set are established.
And step S120, constructing a YOLO network which comprises a target detection network and a double attention channel, and obtaining parameters of the network through later training so that the network can detect the ship, the ship name and the flag in the video.
And S130, training the constructed neural network by using the ship target detection data set and the water traffic violation behavior data set to obtain parameters of the network structure.
And step S140, inputting the channel monitoring video to be detected into a ship target detection model to obtain a local image of the detected target ship.
And S150, inputting the detected target ship area image into the water traffic violation detection network model, and outputting the detection result of the violation. The final display results are shown in fig. 2.
The technical means disclosed in the invention scheme are not limited to the technical means disclosed in the above embodiments, but also include the technical scheme formed by any combination of the above technical features. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and such improvements and modifications are also considered to be within the scope of the present invention.
Claims (6)
1. A method for detecting water traffic violation behaviors based on a YOLO target detection algorithm is characterized by comprising the following steps:
(1) the method comprises the steps that a channel ship video shot by a monitoring camera is subjected to frame extraction, and a ship, a ship name and a flag in the channel ship video are marked to serve as a data set;
(2) establishing a YOLO network structure;
(3) training a YOLO network structure by using the established data set, and adjusting network parameters so as to obtain a ship target detection model and a water traffic violation behavior detection model;
(4) inputting the monitoring video into a ship target detection model, and detecting the position of a target ship in a channel;
(5) and inputting the image of the target ship area into the water traffic violation behavior detection model to obtain a violation behavior detection result of the target ship.
2. The method for detecting the water traffic violation based on the YOLO target detection algorithm as claimed in claim 1, wherein the step (1) is specifically as follows: sampling, taking frames and marking the shot channel monitoring video, and establishing a ship target detection data set and a water traffic violation behavior data set.
3. The method for detecting the water traffic violation based on the YOLO target detection algorithm as claimed in claim 2, wherein the step (2) is specifically as follows: and constructing a YOLO network which comprises a target detection network and a double attention channel, and obtaining parameters of the network through later training so that the network can detect the names and flags of ships and boats in the video.
4. The method for detecting the water traffic violation based on the YOLO target detection algorithm as claimed in claim 3, wherein the step (3) is specifically as follows: and training the constructed neural network by using the ship target detection data set and the water traffic violation behavior data set to obtain parameters of the network structure.
5. The method for detecting the water traffic violation based on the YOLO target detection algorithm as claimed in claim 4, wherein the step (4) is specifically as follows: and inputting the channel monitoring video to be detected into a ship target detection model to obtain a local image of the detected target ship.
6. The method for detecting the water traffic violation based on the YOLO target detection algorithm as claimed in claim 5, wherein the step (5) is specifically as follows: and inputting the detected regional image of the target ship into the water traffic violation detection network model, and outputting the detection result of the violation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110769368.4A CN113657151A (en) | 2021-07-07 | 2021-07-07 | Water traffic violation detection method based on YOLO target detection algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110769368.4A CN113657151A (en) | 2021-07-07 | 2021-07-07 | Water traffic violation detection method based on YOLO target detection algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113657151A true CN113657151A (en) | 2021-11-16 |
Family
ID=78489176
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110769368.4A Pending CN113657151A (en) | 2021-07-07 | 2021-07-07 | Water traffic violation detection method based on YOLO target detection algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113657151A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114429577A (en) * | 2022-01-27 | 2022-05-03 | 西安交通大学 | Flag detection method, system and equipment based on high beacon strategy |
CN114708560A (en) * | 2022-06-06 | 2022-07-05 | 科大天工智能装备技术(天津)有限公司 | YOLOX algorithm-based illegal parking detection method and system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110610165A (en) * | 2019-09-18 | 2019-12-24 | 上海海事大学 | Ship behavior analysis method based on YOLO model |
CN111985363A (en) * | 2020-08-06 | 2020-11-24 | 武汉理工大学 | Ship name recognition system and method based on deep learning framework |
CN112819068A (en) * | 2021-01-29 | 2021-05-18 | 南京长江油运有限公司 | Deep learning-based real-time detection method for ship operation violation behaviors |
-
2021
- 2021-07-07 CN CN202110769368.4A patent/CN113657151A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110610165A (en) * | 2019-09-18 | 2019-12-24 | 上海海事大学 | Ship behavior analysis method based on YOLO model |
CN111985363A (en) * | 2020-08-06 | 2020-11-24 | 武汉理工大学 | Ship name recognition system and method based on deep learning framework |
CN112819068A (en) * | 2021-01-29 | 2021-05-18 | 南京长江油运有限公司 | Deep learning-based real-time detection method for ship operation violation behaviors |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114429577A (en) * | 2022-01-27 | 2022-05-03 | 西安交通大学 | Flag detection method, system and equipment based on high beacon strategy |
CN114429577B (en) * | 2022-01-27 | 2024-03-08 | 西安交通大学 | Flag detection method, system and equipment based on high confidence labeling strategy |
CN114708560A (en) * | 2022-06-06 | 2022-07-05 | 科大天工智能装备技术(天津)有限公司 | YOLOX algorithm-based illegal parking detection method and system |
CN114708560B (en) * | 2022-06-06 | 2022-08-09 | 科大天工智能装备技术(天津)有限公司 | YOLOX algorithm-based illegal parking detection method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109800629B (en) | Remote sensing image target detection method based on convolutional neural network | |
CN113657151A (en) | Water traffic violation detection method based on YOLO target detection algorithm | |
Bedruz et al. | Real-time vehicle detection and tracking using a mean-shift based blob analysis and tracking approach | |
CN108010242B (en) | Security alarm method, system and storage medium based on video identification | |
CN109190455B (en) | Black smoke vehicle identification method based on Gaussian mixture and autoregressive moving average model | |
CN111582182B (en) | Ship name recognition method, system, computer equipment and storage medium | |
Luo et al. | Stationary vehicle detection in aerial surveillance with a UAV | |
CN110969213A (en) | Ship detection method and device based on fast RCNN and electronic equipment | |
CN109271904A (en) | A kind of black smoke vehicle detection method based on pixel adaptivenon-uniform sampling and Bayesian model | |
CN110728212B (en) | Road well lid monitoring device and monitoring method based on computer vision | |
CN112232269B (en) | Ship identity intelligent recognition method and system based on twin network | |
CN111626104B (en) | Cable hidden trouble point detection method and device based on unmanned aerial vehicle infrared thermal image | |
CN116342644A (en) | Intelligent monitoring method and system suitable for coal yard | |
CN115410154A (en) | Method for identifying thermal fault of electrical equipment of wind power engine room | |
CN115331238A (en) | Multi-element fusion ship identity identification method | |
CN114926456A (en) | Rail foreign matter detection method based on semi-automatic labeling and improved deep learning | |
CN113743304A (en) | Moving object detection and identification method for video monitoring | |
Ren et al. | Implementation of vehicle and license plate detection on embedded platform | |
CN110766714A (en) | Automatic segmentation method for water surface floater based on Gaussian mixture model | |
CN114842353B (en) | Neural network remote sensing image target detection method based on self-adaptive target direction | |
CN116503767B (en) | River course floater recognition system based on semantic image processing | |
CN112001320B (en) | Gate detection method based on video | |
CN116416200B (en) | Forest fire smoke detection method based on region matching | |
CN111523492B (en) | Detection method of black smoke vehicle | |
CN117690086A (en) | Flood prevention gate identification and control method and system based on 5G and AI technology |
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 |