CN114937195A - Water surface floating object target detection system based on unmanned aerial vehicle aerial photography and improved YOLO v3 - Google Patents
Water surface floating object target detection system based on unmanned aerial vehicle aerial photography and improved YOLO v3 Download PDFInfo
- Publication number
- CN114937195A CN114937195A CN202210324548.6A CN202210324548A CN114937195A CN 114937195 A CN114937195 A CN 114937195A CN 202210324548 A CN202210324548 A CN 202210324548A CN 114937195 A CN114937195 A CN 114937195A
- Authority
- CN
- China
- Prior art keywords
- aerial vehicle
- unmanned aerial
- water surface
- image
- 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 58
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 40
- 238000007667 floating Methods 0.000 title claims abstract description 20
- 238000013528 artificial neural network Methods 0.000 claims abstract description 16
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 claims abstract description 12
- 238000005259 measurement Methods 0.000 claims abstract description 12
- 230000005540 biological transmission Effects 0.000 claims abstract description 9
- 238000004458 analytical method Methods 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 230000001133 acceleration Effects 0.000 claims abstract description 4
- 238000004891 communication Methods 0.000 claims description 10
- 238000000034 method Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 4
- 238000013500 data storage Methods 0.000 claims description 3
- 238000003064 k means clustering Methods 0.000 claims description 3
- 208000034699 Vitreous floaters Diseases 0.000 abstract description 14
- 238000005516 engineering process Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 239000008267 milk Substances 0.000 description 2
- 210000004080 milk Anatomy 0.000 description 2
- 235000013336 milk Nutrition 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004873 anchoring Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Probability & Statistics with Applications (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a water surface floater target detection system based on unmanned aerial vehicle aerial photography and improved YOLOv3, wherein an unmanned aerial vehicle flight control system comprises an unmanned aerial vehicle attitude measurement system and a main controller, the unmanned aerial vehicle attitude measurement system is composed of an acceleration sensor and a gyroscope, and the main controller is composed of attitude resolving and control calculation; the image acquisition system consists of an image detection system, an image shooting system and an image transmission system, wherein the image detection system selects an improved YOLOv3 neural network as a detection model; the software system is written to an interface of a PC (personal computer) end through pythonqt, and a neural network is deployed on the PC end, the software system is transmitted to the PC end through an unmanned aerial vehicle aerial photograph for analysis, so that the problems that the coverage range of water surface floaters is large, the moving capability is strong, and the types and the quantity of the whole river surface floaters cannot be detected in real time manually can be effectively solved; meanwhile, the system can be better adapted to the detection of the water surface floating object by adopting the improved YOLOv3 neural network.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a water surface floating object target detection system based on unmanned aerial vehicle aerial photography and improved YOLO v 3.
Background
The water surface floater is an important factor influencing the water quality of the lake channel and the river channel, and the water surface environment can be effectively protected by regularly cleaning the harmful water surface floater. At present, the detection of the water surface floaters mainly depends on manual observation, so that the efficiency is low, the speed is low, the coverage range of the water surface floaters is large, the moving capability is strong, and the types and the quantity of the floaters on the whole river surface cannot be detected in real time manually; therefore, the system for detecting the water surface floating object target based on unmanned aerial vehicle aerial photography and improved YOLO v3 is provided, and real-time detection of the whole water surface pollutants in the river channel is realized by transmitting an aerial image of the unmanned aerial vehicle to a PC (personal computer) terminal for data visualization based on the improved YOLO v 3.
Disclosure of Invention
The present invention aims to solve the above problems of the prior art by providing a system for detecting a floating object on water surface based on unmanned aerial vehicle aerial photography and improved YOLO v 3.
In order to achieve the purpose, the invention provides the following technical scheme: the system for detecting the water surface floating object target based on unmanned aerial vehicle aerial photography and improved YOLO v3 comprises an unmanned aerial vehicle flight control system, an image acquisition system and a software system, wherein the unmanned aerial vehicle flight control system comprises an unmanned aerial vehicle attitude measurement system and a main controller, the unmanned aerial vehicle attitude measurement system is composed of an acceleration sensor and a gyroscope, and the main controller is composed of attitude resolving and control calculation; the image acquisition system consists of an image detection system, an image shooting system and an image transmission system, wherein the image detection system selects an improved YOLO v3 neural network as a detection model; the software system is an interface written to the PC end through pythonqt, the neural network is deployed on the PC end, and the software system is connected with the image acquisition system through the wireless communication module.
As a preferred technical scheme of the invention, the main controller is connected with the remote controller through the wireless communication module, the main controller is electrically connected with the driving motor, and the unmanned aerial vehicle attitude measurement system is connected with the main controller.
As a preferred technical scheme of the invention, the hardware of the unmanned aerial vehicle flight control system comprises an unmanned aerial vehicle and a panoramic camera, and the panoramic camera is installed on the bottom of the unmanned aerial vehicle.
As a preferred technical solution of the present invention, the image capturing system captures an article detected by the image detecting system through a camera, and the image transmission system transmits an image by using a bluetooth or 5G module.
As a preferred technical scheme of the invention, the improvement steps of the YOLO v3 are as follows:
s1: the selection of the anchor box is improved by using a kmeans + + algorithm, and the problem that different clustering results are caused by the fact that different K values need to be input manually in the existing K-mean algorithm is solved by using the kmeans + + algorithm;
s2: the weight of small target detection is improved, and the detection precision of the small target detection is improved;
L′ IOC (x，y，w，h)＝(2-wh) 2 L IOC (x，y，w，h)
s3: increasing the coefficients of a position loss function and a classification loss function to enable YOLO v3 to be better adapted to multi-class target detection of water surface floating objects, and obtaining a proper weight coefficient through continuously adjusting the coefficients;
L loss ＝0.5l ob +1.6l lo +0.9l cl ，
s4: the CIOU algorithm is adopted to improve that the IOU is not sensitive to the size of a target object and cannot reflect the distance problem when a detection frame and a real frame are not overlapped any more;
s5: measuring the precision and speed of YOLO v3 by adopting mAP rate and FPS;
as a preferred technical solution of the present invention, the improved algorithm in S1 includes the following specific steps:
s11: inputting data;
s12: randomly selecting a clustering center;
s13: calculating the distance D (x) between all the point clustering centers, and selecting a new clustering center according to D (x);
s14: whether the clustering of the two clustering centers meets the requirement or not, if not, returning to S13, and if so, storing the clustering centers;
s15: and judging whether the total number of the clustering centers is equal to K, if not, returning to S12 to randomly select the clustering centers, and if so, calculating by a K-means clustering analysis algorithm.
As a preferred technical scheme of the invention, the flight shooting and target detection method of the unmanned aerial vehicle comprises the following specific steps:
step 1: the unmanned aerial vehicle is prevented from being on the horizontal plane, and the unmanned aerial vehicle is controlled to start to lift off through a remote controller;
step 2: the method comprises the steps that aerial photography is conducted on water surface floating objects through a camera on a remote control unmanned aerial vehicle;
and 3, step 3: the image is transmitted to the PC end through the wireless communication module, utilizes the PC end to handle picture information in real time, then carries out data storage, and unmanned aerial vehicle lasts the flight simultaneously, just can return to the journey automatically until receiving the information of returning the journey.
The invention has the beneficial effects that: the unmanned aerial vehicle aerial photography image is transmitted to the PC terminal for analysis, so that the problems that the coverage range of the water surface floaters is large, the moving capability is strong, and the types and the quantity of the whole river surface floaters cannot be detected in real time manually can be effectively solved; meanwhile, the system can be better adapted to the detection of the water surface floating object by adopting an improved YOLO v3 neural network; meanwhile, the system can effectively and scientifically further analyze the overall damage of the water surface floating objects to the river by detecting 17 types of ships, plastic bottles, milk boxes, plastic bottles and the like.
Drawings
FIG. 1 is a flow chart of the kmeans + + algorithm of the present invention;
FIG. 2 is a diagram of the neural network architecture of the YOLO v3 of the present invention;
FIG. 3 is a block diagram of an unmanned aerial vehicle flight control system of the present invention;
FIG. 4 is a flow chart of the detection method of the present invention;
FIG. 5 is a data chart of the experimental results of example 1 of the present invention.
Detailed Description
The following detailed description of the preferred embodiments of the present invention, taken in conjunction with the accompanying drawings, will make the advantages and features of the invention more readily understood by those skilled in the art, and thus will more clearly and distinctly define the scope of the invention.
Referring to fig. 1-4, the present invention provides a technical solution: the system for detecting the water surface floating object target based on unmanned aerial vehicle aerial photography and improved YOLO v3 comprises an unmanned aerial vehicle flight control system, an image acquisition system and a software system, wherein the unmanned aerial vehicle flight control system comprises an unmanned aerial vehicle attitude measurement system and a main controller, the unmanned aerial vehicle attitude measurement system is composed of an acceleration sensor and a gyroscope, and the main controller is composed of attitude resolving and control calculation; the image acquisition system consists of an image detection system, an image shooting system and an image transmission system, wherein the image detection system selects an improved YOLO v3 neural network as a detection model; the software system is an interface which is written to the PC end through pythonqt, the neural network is deployed on the PC end, and the software system is connected with the image acquisition system through the wireless communication module.
The main controller is connected with the remote controller through the wireless communication module, the main controller is electrically connected with the driving motor, and the unmanned aerial vehicle attitude measurement system is connected with the main controller; the hardware of the unmanned aerial vehicle flight control system comprises an unmanned aerial vehicle and a panoramic camera, and the panoramic camera is installed on the bottom of the unmanned aerial vehicle; the image shooting system shoots the object detected by the image detection system through the camera, and the image transmission system adopts a Bluetooth or 5G module for image transmission.
The modification procedure for YOLO v3 was as follows:
s1: the method has the advantages that the selection of the anchor box is improved by means of a kmeans + + algorithm, and the problem that different clustering results are caused by the fact that different K values need to be input manually in the existing K-mean algorithm is solved by means of the kmeans + + algorithm;
YOLOv3 uses an anchoring mechanism by borrowing the idea of Fast-RCNN to improve the detection speed and reduce the difficulty in training, but the K-mean algorithm K value selection is sensitive to the setting of an initial value and needs manual input of the number of classes, but different artificial input of K values can lead to different clustering results, so the problem is improved by using a kmeans + + algorithm;
s2: the weight of small target detection is improved, and the detection precision of the small target detection is improved;
L′ IOC (x，y，w，h)＝(2-wh) 2 L IOC (x，y，w，h)
poor accuracy of neural network target detection of small targets is a common problem, and it has two main reasons: firstly, the deep characteristic layer has low resolution, the small target has small volume, and the condition of missing detection is easy to occur; the number of small targets is small, so that the condition of unbalanced samples during training is easily caused;
s3: increasing the position loss function and the classification loss function coefficient to enable YOLO v3 to be better adapted to the detection of multiple classes of targets such as water surface floating objects, and obtaining a proper weight coefficient through the continuously adjusted coefficient;
L loss ＝0.5l ob +1.6l lo +0.9l cl ，
the loss function error of Yolov3 is bounded by box coordinate regression error, confidence error, and classification error; for the original 1: 1: 1 proportion of weight distribution; therefore, the position loss function and the classification loss function coefficient are increased, so that the YOLO v3 is better adapted to the detection of multiple classes of targets such as water surface floating objects, and a more appropriate weight coefficient is obtained by continuous coefficient adjustment;
s4: the CIOU algorithm is adopted to improve that the IOU is not sensitive to the size of a target object and cannot reflect the distance problem when a detection frame and a real frame are not overlapped any more;
s5: measuring the precision and speed of YOLO v3 by adopting mAP rate and FPS;
the IOU is a standard for measuring the accuracy of detecting a corresponding object in a specific data set, but the IOU is not sensitive to the size of a target object and cannot reflect the distance problem when a detection frame and a real frame are not overlapped any more, so that the problem can be effectively solved by using a CIOU algorithm to improve the IOU, wherein the CIOU comprehensively considers the overlapping area, the distance of a central point and the aspect ratio.
The specific steps of the improved algorithm in the step S1 are as follows:
s11: inputting data;
s12: randomly selecting a clustering center;
s13: calculating the distance D (x) between all the point clustering centers, and selecting a new clustering center according to D (x);
s14: whether the clustering of the two clustering centers meets the requirement or not is judged, if not, the clustering returns to S13, and if so, the clustering centers are stored;
s15: and judging whether the total number of the clustering centers is equal to K, if not, returning to S12 to randomly select the clustering centers, and if so, calculating by a K-means clustering analysis algorithm.
A water surface floating object target detection method based on unmanned aerial vehicle aerial photography and improved YOLO v3 comprises the following specific steps:
step 1: the unmanned aerial vehicle is prevented from being on the horizontal plane, and the unmanned aerial vehicle is controlled to start to lift off through a remote controller;
step 2: the camera on the unmanned aerial vehicle is remotely controlled to take aerial photos of the floating objects on the water surface;
and step 3: the image is transmitted to the PC end through the wireless communication module, utilizes the PC end to handle picture information in real time, then carries out data storage, and unmanned aerial vehicle lasts the flight simultaneously, just can return to the journey automatically until receiving the information of returning the journey.
The system combines a machine vision technology, an image processing technology and an unmanned aerial vehicle flight control technology, and simultaneously, a software design technology is also applied; the method comprises the steps of realizing image acquisition, image processing and detection data summarization of the water surface floater by building an image acquisition hardware platform and a designed software system; and the YOLO v3 neural network is adopted, so that the method has the following advantages:
darknet53 is a backbone network of YOLOv 3; darknet53 removes the public pooling layer and the full-link layer of the convolutional neural network, and only uses the convolutional layer to extract the features; the detection speed of YOLOv3 is greatly increased;
the residual layer is added to the Darknet53-network, and the residual structure in the ResNet can effectively solve the degradation problem caused by the network depth;
3, using a similar characteristic pyramid method to solve the multi-scale problem of target detection by using the FPN concept, YOLOv3 to achieve a better small target detection effect; by using an up-sampling fusion method, three different-size scale feature maps are respectively fused to generate a multi-scale feature map, so that the detection precision is improved.
Example 1: carrying out experimental operation by the following steps;
step 1: the unmanned aerial vehicle is prevented from being on the horizontal plane, and the unmanned aerial vehicle is controlled to start to lift off through a remote controller;
step 2: the method comprises the steps that aerial photography is conducted on water surface floating objects through a camera on a remote control unmanned aerial vehicle;
and step 3: the image is transmitted to the PC end through the wireless communication module, the image information is processed in real time by the PC end, then data is stored, meanwhile, the unmanned aerial vehicle continuously flies, and the automatic return flight is not carried out until the return flight information is received;
an improved YOLO v3 neural network is employed within an image processing system, as follows:
1. the selection of the anchor box is improved by using a kmeans + + algorithm, and the clustering accuracy reaches 70.56%; the anchor box for kmeans + + algorithm clustering is shown in table 1;
TABLE II
ANCHOR BOX SIZE
TABLE 1
2. Neural network training parameters for YOLO v3 are shown in table 2:
TABLE III
SPECIFIC PARAMETERS OF NETWORK TRAINING
TABLE 2
3. Measuring the precision and speed of YOLOv3 through mAP rate and FPS;
as can be seen from table 3 and fig. 5, experiments prove that the improved YOLOv3 algorithm is improved by 5.25% compared with the original algorithm by the mAP rate, and reaches 84.81%;
TABLE IV
TEST RESULT
table 3.
The unmanned aerial vehicle aerial photography image is transmitted to the PC terminal for analysis, so that the problems that the coverage range of the water surface floaters is large, the moving capability is strong, and the types and the quantity of the whole river surface floaters cannot be detected in real time manually can be effectively solved; meanwhile, the system can be better adapted to the detection of the water surface floater target by adopting an improved YOLO v3 neural network; meanwhile, the system can effectively and scientifically further analyze the overall damage of the water surface floating objects to the river by detecting 17 types of ships, plastic bottles, milk boxes, plastic bottles and the like.
The above examples only show some embodiments of the present invention, and the description thereof is specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention.
Claims (7)
1. Water surface floater target detection system based on unmanned aerial vehicle takes photo by plane and modified YOLO v3, including unmanned aerial vehicle flight control system, image acquisition system and software system, its characterized in that: the unmanned aerial vehicle flight control system comprises an unmanned aerial vehicle attitude measurement system and a main controller, wherein the unmanned aerial vehicle attitude measurement system is composed of an acceleration sensor and a gyroscope, and the main controller is composed of attitude calculation and control calculation; the image acquisition system consists of an image detection system, an image shooting system and an image transmission system, wherein the image detection system selects an improved YOLO v3 neural network as a detection model; the software system is an interface written to the PC end through pythonqt, the neural network is deployed on the PC end, and the software system is connected with the image acquisition system through the wireless communication module.
2. The drone aerial and modified YOLO v 3-based water surface float target detection system of claim 1, wherein: the unmanned aerial vehicle attitude measurement system comprises a main controller, a driving motor, a wireless communication module, a remote controller, a driving motor and an unmanned aerial vehicle attitude measurement system.
3. The drone aerial and modified YOLO v 3-based water surface float target detection system of claim 1, wherein: the hardware of the unmanned aerial vehicle flight control system comprises an unmanned aerial vehicle and a panoramic camera, and the panoramic camera is installed on the bottom of the unmanned aerial vehicle.
4. The drone-aerial and modified YOLO v 3-based water surface float target detection system of claim 1, wherein: the image shooting system shoots the object detected by the image detection system through the camera, and the image transmission system adopts a Bluetooth or 5G module for image transmission.
5. The drone-aerial and modified YOLO v 3-based water surface float target detection system of claim 1, wherein: the improvement of the YOLO v3 is as follows:
s1: the selection of the anchor box is improved by using a kmeans + + algorithm, and the problem that different clustering results are caused by the fact that different K values need to be input manually in the existing K-mean algorithm is solved by using the kmeans + + algorithm;
s2: the weight of small target detection is improved, and the detection precision of the small target detection is improved;
L′ IOC (x，y，w，h)＝(2-wh) 2 L IOC (x，y，w，h)
s3: increasing the position loss function and the classification loss function coefficient to enable YOLO v3 to be better adapted to the detection of multiple classes of targets such as water surface floating objects, and obtaining a proper weight coefficient through the continuously adjusted coefficient;
L loss ＝0.5l ob +1.6l lo +0.9l cl ，；
s4: the CIOU algorithm is adopted to improve that the IOU is not sensitive to the size of a target object and cannot reflect the distance problem when a detection frame and a real frame are not overlapped any more;
s5: measuring the precision and speed of YOLO v3 by adopting mAP rate and FPS;
6. the drone-aerial and modified YOLO v 3-based water surface float target detection system of claim 5, wherein: the specific steps of the improved algorithm in S1 are as follows:
s11: inputting data;
s12: randomly selecting a clustering center;
s13: calculating the distance D (x) between all the point clustering centers, and selecting a new clustering center according to the distance D (x);
s14: whether the clustering of the two clustering centers meets the requirement or not is judged, if not, the clustering returns to S13, and if so, the clustering centers are stored;
s15: and judging whether the total number of the clustering centers is equal to K, if not, returning to S12 to randomly select the clustering centers, and if so, calculating by a K-means clustering analysis algorithm.
7. The drone-aerial and modified YOLO v 3-based water surface float target detection system of claim 1, wherein: the flight shooting and target detection method of the unmanned aerial vehicle comprises the following specific steps:
step 1: the unmanned aerial vehicle is prevented from being on the horizontal plane, and the unmanned aerial vehicle is controlled to start to lift off through a remote controller;
and 2, step: the method comprises the steps that aerial photography is conducted on water surface floating objects through a camera on a remote control unmanned aerial vehicle;
and step 3: the image is transmitted to the PC end through the wireless communication module, utilizes the PC end real-time processing picture information, then data storage, and unmanned aerial vehicle lasts the flight simultaneously, just can return to the journey automatically until receiving the information of returning to the journey.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210324548.6A CN114937195A (en) | 2022-03-29 | 2022-03-29 | Water surface floating object target detection system based on unmanned aerial vehicle aerial photography and improved YOLO v3 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210324548.6A CN114937195A (en) | 2022-03-29 | 2022-03-29 | Water surface floating object target detection system based on unmanned aerial vehicle aerial photography and improved YOLO v3 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114937195A true CN114937195A (en) | 2022-08-23 |
Family
ID=82862263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210324548.6A Pending CN114937195A (en) | 2022-03-29 | 2022-03-29 | Water surface floating object target detection system based on unmanned aerial vehicle aerial photography and improved YOLO v3 |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114937195A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111079540A (en) * | 2019-11-19 | 2020-04-28 | 北航航空航天产业研究院丹阳有限公司 | Target characteristic-based layered reconfigurable vehicle-mounted video target detection method |
| CN111832379A (en) * | 2019-10-15 | 2020-10-27 | 中国石油化工股份有限公司 | Unmanned aerial vehicle real-time video detection system based on convolutional neural network |
| CN111898501A (en) * | 2020-07-17 | 2020-11-06 | 东南大学 | Unmanned aerial vehicle online aerial photography vehicle identification and statistics method for congested road sections |
| CN111898651A (en) * | 2020-07-10 | 2020-11-06 | 江苏科技大学 | Tree detection method based on Tiny Yolov3 algorithm |
| CN112613343A (en) * | 2020-12-01 | 2021-04-06 | 浙江大学 | Improved YOLOv 4-based river waste monitoring method |
| CN113344885A (en) * | 2021-06-15 | 2021-09-03 | 温州大学 | River floating object detection method based on cascade convolution neural network |
| CN113435269A (en) * | 2021-06-10 | 2021-09-24 | 华东师范大学 | Improved water surface floating object detection and identification method and system based on YOLOv3 |
| CN113920400A (en) * | 2021-10-14 | 2022-01-11 | 辽宁工程技术大学 | Metal surface defect detection method based on improved YOLOv3 |
| CN114037918A (en) * | 2021-11-10 | 2022-02-11 | 吉林大学 | Photovoltaic hot spot detection method based on unmanned aerial vehicle inspection and image processing |
-
2022
- 2022-03-29 CN CN202210324548.6A patent/CN114937195A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111832379A (en) * | 2019-10-15 | 2020-10-27 | 中国石油化工股份有限公司 | Unmanned aerial vehicle real-time video detection system based on convolutional neural network |
| CN111079540A (en) * | 2019-11-19 | 2020-04-28 | 北航航空航天产业研究院丹阳有限公司 | Target characteristic-based layered reconfigurable vehicle-mounted video target detection method |
| CN111898651A (en) * | 2020-07-10 | 2020-11-06 | 江苏科技大学 | Tree detection method based on Tiny Yolov3 algorithm |
| CN111898501A (en) * | 2020-07-17 | 2020-11-06 | 东南大学 | Unmanned aerial vehicle online aerial photography vehicle identification and statistics method for congested road sections |
| CN112613343A (en) * | 2020-12-01 | 2021-04-06 | 浙江大学 | Improved YOLOv 4-based river waste monitoring method |
| CN113435269A (en) * | 2021-06-10 | 2021-09-24 | 华东师范大学 | Improved water surface floating object detection and identification method and system based on YOLOv3 |
| CN113344885A (en) * | 2021-06-15 | 2021-09-03 | 温州大学 | River floating object detection method based on cascade convolution neural network |
| CN113920400A (en) * | 2021-10-14 | 2022-01-11 | 辽宁工程技术大学 | Metal surface defect detection method based on improved YOLOv3 |
| CN114037918A (en) * | 2021-11-10 | 2022-02-11 | 吉林大学 | Photovoltaic hot spot detection method based on unmanned aerial vehicle inspection and image processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113038016B (en) | Unmanned aerial vehicle image acquisition method and unmanned aerial vehicle | |
| CN107203222B (en) | Method for realizing ship water gauge image and video shooting by unmanned aerial vehicle | |
| JP6496323B2 (en) | System and method for detecting and tracking movable objects | |
| US20220301302A1 (en) | Air and sea based fishing data collection and analysis systems and methods | |
| US11900668B2 (en) | System and method for identifying an object in water | |
| US20170102699A1 (en) | Drone control through imagery | |
| TW201925033A (en) | Unmanned aerial vehicle, control system for unmanned aerial vehicle and control method thereof | |
| WO2019000325A1 (en) | Augmented reality method for aerial photography of unmanned aerial vehicle, processor, and unmanned aerial vehicle | |
| CN111765974B (en) | Wild animal observation system and method based on miniature refrigeration thermal infrared imager | |
| KR101261409B1 (en) | System for recognizing road markings of image | |
| CN109145747A (en) | A kind of water surface panoramic picture semantic segmentation method | |
| JPWO2019082301A1 (en) | Detection system, detection method, and program | |
| CN110127000B (en) | Intelligent navigation eye system for transport ship | |
| CN111814721B (en) | Airfield runway foreign matter detection and classification method based on unmanned aerial vehicle high-low altitude combined scanning | |
| CN107343165A (en) | A kind of monitoring method, equipment and system | |
| CN115793689A (en) | Unmanned aerial vehicle automatic overhead transmission line inspection method and system based on front-end target identification | |
| CN114693661A (en) | Rapid sorting method based on deep learning | |
| CN112001226A (en) | Unmanned 3D target detection method and device and storage medium | |
| CN113408510B (en) | Transmission line target deviation rectifying method and system based on deep learning and one-hot coding | |
| CN107221006A (en) | A kind of communication single pipe tower slant detection method based on unmanned plane imaging platform | |
| JP2019185689A (en) | System for specifying detection object position | |
| CN113822217A (en) | Ship tail gas monitoring method based on AIS and video image analysis | |
| CN110376593B (en) | Target sensing method and device based on laser radar | |
| WO2019189381A1 (en) | Moving body, control device, and control program | |
| CN114937195A (en) | Water surface floating object target detection system based on unmanned aerial vehicle aerial photography and improved YOLO v3 |
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: 20220823 |
|
| RJ01 | Rejection of invention patent application after publication |