CN112633375A - Bird detection method and device, computer equipment and storage medium - Google Patents
Bird detection method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN112633375A CN112633375A CN202011540324.6A CN202011540324A CN112633375A CN 112633375 A CN112633375 A CN 112633375A CN 202011540324 A CN202011540324 A CN 202011540324A CN 112633375 A CN112633375 A CN 112633375A
- Authority
- CN
- China
- Prior art keywords
- bird
- detection
- image
- model
- detection result
- 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 173
- 238000012549 training Methods 0.000 claims abstract description 51
- 230000001846 repelling effect Effects 0.000 claims abstract description 42
- 238000013135 deep learning Methods 0.000 claims abstract description 30
- 238000004590 computer program Methods 0.000 claims description 21
- 238000000034 method Methods 0.000 claims description 20
- 238000004458 analytical method Methods 0.000 claims description 17
- 238000007619 statistical method Methods 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 6
- 238000011897 real-time detection Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 9
- 230000008859 change Effects 0.000 description 7
- 238000012360 testing method Methods 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 230000036544 posture Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M29/00—Scaring or repelling devices, e.g. bird-scaring apparatus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Birds (AREA)
- Insects & Arthropods (AREA)
- Pest Control & Pesticides (AREA)
- Wood Science & Technology (AREA)
- Zoology (AREA)
- Environmental Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a bird detection method, a bird detection device, computer equipment and a storage medium, wherein the bird detection method comprises the steps of obtaining an image to be detected; inputting the image to be detected into a bird detection model for detection to obtain a detection result; feeding back the detection result; the bird detection model is obtained by adopting an image with a bird label as a sample set to train a deep learning network for the second time. According to the bird detection method, after the image to be detected is obtained, the bird condition in the image is detected through the bird detection model, when the detection result shows that the bird is close to the camera, subsequent bird repelling and other operations can be carried out according to the fed-back detection result, when the detection result shows that the bird is a small target bird, the bird repelling operation can be carried out according to the subsequent real-time detection result, through secondary training of the bird detection model, the birds with short distance and large pixels can be detected accurately, the birds with long distance and small targets can be detected accurately, and the accuracy is high.
Description
Technical Field
The invention relates to a bird detection method, in particular to a bird detection method, a bird detection device, computer equipment and a storage medium.
Background
At present, under a single camera, different kinds of birds have different distances from the camera, so that the characteristic quantity of the birds appearing in an image is different, meanwhile, the flying postures of the birds are different, the detection difficulty of a bird target detection algorithm is increased, the detection rate of the birds for long distances is lower and lower, however, in practical application, the detection target distance required is often relatively far, the long-distance target detection has more economic benefits, but the current bird detection method cannot accurately identify the birds with small targets or can identify the birds with small targets, but the accuracy is not high.
Therefore, it is necessary to design a new method to accurately detect birds with large pixels in a short distance and to accurately detect birds with small targets in a long distance, and the accuracy is high.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a bird detection method, a bird detection device, computer equipment and a storage medium.
In order to achieve the purpose, the invention adopts the following technical scheme: a method of bird detection comprising:
acquiring an image to be detected;
inputting the image to be detected into a bird detection model for detection to obtain a detection result;
feeding back the detection result;
the bird detection model is obtained by adopting an image with a bird label as a sample set to train a deep learning network for the second time.
The further technical scheme is as follows: the bird detection model is obtained by adopting an image with a bird label as a sample set to train a deep learning network for the second time, and comprises the following steps:
acquiring an image with a bird tag to obtain a sample set;
constructing a deep learning network;
carrying out coarse precision training on the deep learning network by using a sample set to obtain an intermediate model;
and carrying out high-precision training on the intermediate model to obtain a bird detection model.
The further technical scheme is as follows: the sample set is formed by extracting data belonging to birds from a public data set and marking the bird data shot by a camera.
The further technical scheme is as follows: right the intermediate model carries out high accuracy training to obtain birds detection model, include:
and training the intermediate model by utilizing bird data shot by the camera in the sample set to obtain a bird detection model.
The further technical scheme is as follows: after the feedback of the detection result, the method further comprises:
and tracking the corresponding birds according to the detection result, and performing bird repelling operation.
The further technical scheme is as follows: after the corresponding birds are tracked according to the detection result and bird repelling operation is carried out, the method further comprises the following steps:
and establishing a bird condition database according to the detection result, performing statistical analysis on the bird condition database to obtain an analysis result, and feeding back the analysis result to the terminal.
The present invention also provides a bird detection device comprising:
the image acquisition unit is used for acquiring an image to be detected;
the detection unit is used for inputting the image to be detected into a bird detection model for detection so as to obtain a detection result;
and the feedback unit is used for feeding back the detection result.
The further technical scheme is as follows: further comprising:
and the model acquisition unit is used for adopting the image with the bird label as a sample set to secondarily train the deep learning network so as to obtain the bird detection model.
The invention also provides computer equipment which comprises a memory and a processor, wherein the memory is stored with a computer program, and the processor realizes the method when executing the computer program.
The invention also provides a storage medium storing a computer program which, when executed by a processor, is operable to carry out the method as described above.
Compared with the prior art, the invention has the beneficial effects that: according to the bird detection method, after the image to be detected is obtained, the bird condition in the image is detected through the bird detection model, when the detection result shows that the bird is close to the camera, subsequent bird repelling and other operations can be carried out according to the fed-back detection result, when the detection result shows that the bird is a small target bird, the bird repelling operation can be carried out according to the subsequent real-time detection result, through secondary training of the bird detection model, the birds with short distance and large pixels can be detected accurately, the birds with long distance and small targets can be detected accurately, and the accuracy is high.
The invention is further described below with reference to the accompanying drawings and specific embodiments.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic view of an application scenario of the bird detection method according to the embodiment of the present invention;
FIG. 2 is a schematic flow chart of a bird detection method according to an embodiment of the present invention;
FIG. 3 is a schematic view of a sub-flow of a bird detection method according to an embodiment of the present invention;
FIG. 4 is a schematic flow chart of a bird detection method according to another embodiment of the present invention;
FIG. 5 is a schematic flow chart of a bird detection method according to another embodiment of the present invention;
FIG. 6 is a schematic block diagram of a bird detection device provided by embodiments of the present invention;
FIG. 7 is a schematic block diagram of a bird detection device provided in another embodiment of the present invention;
FIG. 8 is a schematic block diagram of a bird detection device provided in another embodiment of the present invention;
FIG. 9 is a schematic block diagram of a computer device provided by an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
Referring to fig. 1 and fig. 2, fig. 1 is a schematic view of an application scenario of the bird detection method according to the embodiment of the present invention. Fig. 2 is a schematic flow chart of a bird detection method provided by an embodiment of the present invention. The bird detection method is applied to a server. The server performs data interaction with the terminal and the camera, acquires an image to be detected through the camera, and performs bird detection by adopting a bird detection model, wherein the bird detection model can detect out birds with small targets and feeds back detection results to the terminal so as to perform analysis or bird repelling operation.
Fig. 2 is a schematic flow chart of a bird detection method provided by an embodiment of the present invention. As shown in fig. 2, the method includes the following steps S110 to S130.
And S110, acquiring an image to be detected.
In this embodiment, the image to be detected is an image captured by a camera at a specific position, and in this embodiment, the image to be detected is an environmental image captured by a camera installed on an airport runway at a specific position of an airport.
And S120, inputting the image to be detected into a bird detection model for detection to obtain a detection result.
In this embodiment, the bird detection model is obtained by secondarily training a deep learning network by using an image with a bird tag as a sample set. The training is divided into two stages, the first stage trains the model with coarse precision, and the second stage trains the model with high precision. And training a target detection algorithm by using a deep learning network framework, setting training parameters, carrying out classification training on labeled data, testing and evaluating the trained model, and selecting the optimal model.
The detection result refers to whether birds exist in the current environment and the specific positions of the birds when the birds exist.
In an embodiment, referring to fig. 3, the bird detection model is obtained by training the deep learning network twice using the image with the bird tag as the sample set, and may include steps S121 to S124.
And S121, acquiring an image with a bird tag to obtain a sample set.
In this embodiment, the sample set is formed by extracting data belonging to birds from a common data set and labeling the data of the birds photographed by a camera. Wherein, a public data set refers to a data set of related platforms on a network, such as hundreds of degrees, relating to birds.
Specifically, the sample set is a data set which divides the birds into three different levels, namely, the three levels are divided according to the size, the data are collected and manually shot on the network, images under various environment backgrounds are collected in some open source databases as much as possible to form the sample set, the sample set is labeled, the labeled content comprises labels formed by the positions of the birds, and meanwhile, the labeled sample set is classified into data according to the pixels occupied by the labeling frames.
In this embodiment, birds are labeled one by using a labelImg labeling tool and a rectangular frame, and the labeled labels are bird, and a corresponding label file in txt format is stored, thereby forming a sample set.
Specifically, the width and height of the labeling frame in the sample set are read, and the area of the labeling frame is calculated. The data is divided by three pixel area level nodes with pixel areas of 40 × 40, 20 × 20, 10 × 10, the label is changed to Lbird if the pixel area >40 × 40, changed to Mbird if 40 × 40> 20 × 20, and changed to sbird if 20 × 20> 10 × 10. The data was then divided into training and test sets at 5:1 on these three levels, respectively.
And S122, constructing a deep learning network.
In this embodiment, an efficientdet network framework is adopted as a deep learning network for training.
And S123, carrying out coarse precision training on the deep learning network by using the sample set to obtain an intermediate model.
In this embodiment, the intermediate model refers to a model obtained by training a deep learning network using public data, and the intermediate model is trained using relatively standard data without considering the influence of external environments such as weather.
Selecting a Pythroch as a deep learning network training platform, using an eficientNet network as a main network, selecting a D0 model, setting the size of an input image to be 512 and setting the size of batch _ size to be 64, and if the video memory overflows, adjusting the size of the batch _ size; since the present embodiment needs to detect birds of three kinds of ranks, num _ classes is set to 3. Fast normalized fusion, namely a Fast regularization method is selected as a weighted feature fusion method, and an open source pre-training database model is used as an initialization parameter of the model. Setting the learning rate lr to 0.1 as the initial learning rate, and setting the input resolution parameter compound _ coef to 0, i.e. the DO model of efficientDet, and corresponding to b0 of efficientNet network. And training the primary version of the model by taking the common data set as a training set. And stopping training after the training observation total loss value, the target recognition loss value and the confidence coefficient loss value tend to be stable, and storing the rough model obtained by training.
And S124, carrying out high-precision training on the intermediate model to obtain a bird detection model.
Specifically, the bird data shot by the camera in the sample set are utilized to train the middle model so as to obtain the bird detection model.
And (3) taking the coarse model as an initialization parameter of the fine model, finely adjusting parameters of the coarse model by using data shot by a camera for training, setting the learning rate to be 0.001, setting the batch _ size to be 128, keeping other parameters unchanged, observing the change of each loss value, and stopping training after the change tends to be stable.
For the test process of the bird detection model, a reliability threshold is set to be 0.75, a cross threshold iou _ threshold is set to be 0.5, a target list obj _ ist is [ 'Lbrid', 'Mbrid', 'Sbrid' ], the test of the test data to the model is judged by using an Average accuracy mAP (mean Average Precision) index, if the AP of Lbrid is less than 0.98, the AP of Mbrid is less than 0.95 and Sbrrid is less than 0.90, the training parameter setting is modified or the data set is added for retraining until the mAP of each category is satisfied.
And S130, feeding back the detection result.
And judging whether to drive the bird repelling device or not according to the level and the number of the detected birds. If Lbridge birds are detected and the representative birds are close to the camera, the bird repelling device can be immediately driven, and if Mbridge birds are detected, the set number exceeds 3, and the bird repelling device is driven to repel birds. And only when the bird of the sbrid level is detected, the set number exceeds 10, and the bird repelling equipment is driven again to achieve the purpose of actively repelling the birds.
Through marking the birds targets of different pixels and carrying out multi-classification training, under a single camera, the bird detection method can detect birds with close-range large pixels at high precision and can detect birds with small targets at remote distances more accurately, robustness and adaptability of an algorithm are improved, meanwhile, a better monitoring function of camera equipment is achieved, the monitoring range is enlarged, the number of monitoring equipment is reduced, and equipment purchasing cost and construction cost are reduced. In addition, by tracking the change of birds and detecting the grade, the bird repelling effect of the bird repelling equipment is analyzed, and the corresponding bird repelling equipment is adopted to repel birds for airport personnel aiming at different seasons and the appearance of different birds, so that the more effective bird repelling effect is improved.
According to the bird detection method, after the image to be detected is obtained, the bird condition in the image is detected through the bird detection model, when the detection result shows that the bird is very close to the camera, the subsequent bird repelling operation and the like can be carried out according to the fed-back detection result, when the detection result shows that the bird is a small target bird, the bird repelling operation can be carried out according to the subsequent real-time detection result, through the secondary training of the bird detection model, the purposes that the birds with close-range large pixels can be detected with high precision, the birds with long-range small targets can be detected more accurately are achieved, and the accuracy is high.
Fig. 4 is a schematic flow chart of a bird detection method according to another embodiment of the present invention. As shown in fig. 4, the bird detection method of the present embodiment includes steps S210 to S240. Steps S210 to S230 are similar to steps S110 to S130 in the above embodiments, and are not described herein again. The added step S240 in the present embodiment is explained in detail below.
S240, tracking the corresponding birds according to the detection result, and performing bird repelling operation.
Specifically, a KCF (Kernel Correlation Filter) algorithm is adopted to track birds in a multi-target mode, the detected grade change of each bird is recorded, after the bird repelling device is driven, if the detected type of one bird is changed into Mcird from Lcird within 10 seconds, and then changed into sbard until the bird flies away from a monitoring range, the bird repelling is judged to be effective; if a bird is always tracked and detected in Lcircd, Mcircd or sbird, and no change trend of the grade label occurs, the bird repelling is failed, and the bird repelling device is ineffective in repelling the birds.
Fig. 5 is a schematic flow chart of a bird detection method according to another embodiment of the present invention. As shown in fig. 5, the bird detection method of the present embodiment includes steps S310 to S350. Steps S310 to S340 are similar to steps S210 to S240 in the above embodiments, and are not described herein again. The added step S350 in the present embodiment is explained in detail below.
S350, establishing a bird condition database according to the detection result, performing statistical analysis on the bird condition database to obtain an analysis result, and feeding back the analysis result to the terminal.
In this embodiment, the bird condition database is a database for recording the bird appearance time, the bird appearance location, and the related images, and the analysis result is the number and the appearance location of various kinds of birds appearing in different seasons.
The bird condition database comprises the time, weather, quantity and type of birds and the effective bird repelling state of the bird repelling device. Through analysis bird situation database, judge the quantity of different seasons birds, the kind drives the bird to carry out the pertinence ecology, through statistical analysis different seasons, different kinds of birds appear, drives bird validity of bird equipment, carries out the change of pertinence and drives bird equipment to increase and drive bird efficiency. The effectiveness of bird repelling equipment is analyzed through big data statistics, and what kind of bird repelling equipment needs to be placed at different time is analyzed, so that the bird repelling effect is increased, and the aim of comprehensively repelling birds by various kinds of equipment is fulfilled. Through bird condition database, the bird species and the bird quantity that different time and different climates appear are analyzed, and then judge the ecological environment of birds, and then carry out the initiative and drive the bird to reduce the probability that birds appear. After the bird repelling device is driven, the activity state of the bird is tracked, namely the type change of the bird is detected, and the bird repelling effect of the bird repelling device on the bird is judged. In addition, the video of the bird is stored as the evidence for the relevant personnel.
Fig. 6 is a schematic block diagram of a bird detection device 300 according to an embodiment of the present invention. As shown in fig. 6, the present invention also provides a bird detection device 300 corresponding to the above bird detection method. The bird detection device 300 includes a unit for performing the bird detection method described above, and the device may be configured in a server. Specifically, referring to fig. 6, the bird detection device 300 includes an image acquisition unit 301, a detection unit 302, and a feedback unit 303.
An image acquisition unit 301 for acquiring an image to be detected; the detection unit 302 is used for inputting the image to be detected into a bird detection model for detection so as to obtain a detection result; a feedback unit 303, configured to feed back the detection result.
In one embodiment, the bird detection device 300 further includes a model acquisition unit.
And the model acquisition unit is used for adopting the image with the bird label as a sample set to secondarily train the deep learning network so as to obtain the bird detection model.
In an embodiment, the model obtaining unit includes a sample set obtaining subunit, a network constructing subunit, a coarse training subunit, and a fine training subunit.
The sample set acquisition subunit is used for acquiring the image with the bird tag to obtain a sample set; the network construction subunit is used for constructing a deep learning network; the rough training subunit is used for carrying out rough precision training on the deep learning network by utilizing the sample set so as to obtain an intermediate model; the accurate training subunit is used for carrying out high-precision training on the intermediate model to obtain a bird detection model, and specifically, the accurate training subunit utilizes bird data shot by a camera in a sample set to train the intermediate model to obtain the bird detection model.
Fig. 7 is a schematic block diagram of a bird detection device 300 according to another embodiment of the present invention. As shown in fig. 7, the bird detection device 300 of the present embodiment is the bird repelling unit 304 added to the above-described embodiment.
And the bird repelling unit 304 is used for tracking the corresponding birds according to the detection result and performing bird repelling operation.
Fig. 8 is a schematic block diagram of a bird detection device 300 according to another embodiment of the present invention. As shown in fig. 8, the bird detection device 300 of the present embodiment is the above-described embodiment with the addition of an analysis unit 305.
The analysis unit 305 is configured to establish a bird condition database according to the detection result, perform statistical analysis on the bird condition database to obtain an analysis result, and feed back the analysis result to the terminal.
It should be noted that, as can be clearly understood by those skilled in the art, the detailed implementation process of the bird detection device 300 and each unit may refer to the corresponding description in the foregoing method embodiment, and for convenience and brevity of description, no further description is provided herein.
Referring to fig. 9, fig. 9 is a schematic block diagram of a computer device according to an embodiment of the present application. The computer device 500 may be a server, wherein the server may be an independent server or a server cluster composed of a plurality of servers.
Referring to fig. 9, the computer device 500 includes a processor 502, memory, and a network interface 505 connected by a system bus 501, where the memory may include a non-volatile storage medium 503 and an internal memory 504.
The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions that, when executed, cause the processor 502 to perform a bird detection method.
The processor 502 is used to provide computing and control capabilities to support the operation of the overall computer device 500.
The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503, and when the computer program 5032 is executed by the processor 502, the processor 502 may be caused to execute a bird detection method.
The network interface 505 is used for network communication with other devices. Those skilled in the art will appreciate that the configuration shown in fig. 9 is a block diagram of only a portion of the configuration associated with the present application and does not constitute a limitation of the computer device 500 to which the present application may be applied, and that a particular computer device 500 may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
Wherein the processor 502 is configured to run the computer program 5032 stored in the memory to implement the following steps:
acquiring an image to be detected; inputting the image to be detected into a bird detection model for detection to obtain a detection result; and feeding back the detection result.
The bird detection model is obtained by adopting an image with a bird label as a sample set to train a deep learning network for the second time.
In an embodiment, when the step of implementing the bird detection model is a step of training a deep learning network secondarily by using an image with a bird tag as a sample set, the processor 502 specifically implements the following steps:
acquiring an image with a bird tag to obtain a sample set; constructing a deep learning network; carrying out coarse precision training on the deep learning network by using a sample set to obtain an intermediate model; and carrying out high-precision training on the intermediate model to obtain a bird detection model.
Wherein, the sample set is formed by extracting data belonging to birds from a public data set and marking the bird data shot by a camera.
In an embodiment, when implementing the step of training the intermediate model with high precision to obtain the bird detection model, the processor 502 specifically implements the following steps:
and training the intermediate model by utilizing bird data shot by the camera in the sample set to obtain a bird detection model.
In an embodiment, after the step of feeding back the detection result, the processor 502 further performs the following steps:
and tracking the corresponding birds according to the detection result, and performing bird repelling operation.
In an embodiment, after the processor 502 tracks the corresponding birds according to the detection result and performs the bird repelling operation, the following steps are further performed:
and establishing a bird condition database according to the detection result, performing statistical analysis on the bird condition database to obtain an analysis result, and feeding back the analysis result to the terminal.
It should be understood that in the embodiment of the present Application, the Processor 502 may be a Central Processing Unit (CPU), and the Processor 502 may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, and the like. Wherein a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
It will be understood by those skilled in the art that all or part of the flow of the method implementing the above embodiments may be implemented by a computer program instructing associated hardware. The computer program includes program instructions, and the computer program may be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the flow steps of the embodiments of the method described above.
Accordingly, the present invention also provides a storage medium. The storage medium may be a computer-readable storage medium. The storage medium stores a computer program, wherein the computer program, when executed by a processor, causes the processor to perform the steps of:
acquiring an image to be detected; inputting the image to be detected into a bird detection model for detection to obtain a detection result; and feeding back the detection result.
The bird detection model is obtained by adopting an image with a bird label as a sample set to train a deep learning network for the second time.
In an embodiment, when the processor executes the computer program to implement the bird detection model as a step of secondarily training a deep learning network by using an image with a bird tag as a sample set, the processor specifically implements the following steps:
acquiring an image with a bird tag to obtain a sample set; constructing a deep learning network; carrying out coarse precision training on the deep learning network by using a sample set to obtain an intermediate model; and carrying out high-precision training on the intermediate model to obtain a bird detection model.
Wherein, the sample set is formed by extracting data belonging to birds from a public data set and marking the bird data shot by a camera.
In an embodiment, when the processor executes the computer program to implement the step of performing high-precision training on the intermediate model to obtain the bird detection model, the processor specifically implements the following steps:
and training the intermediate model by utilizing bird data shot by the camera in the sample set to obtain a bird detection model.
In an embodiment, after the step of feeding back the detection result is realized by the processor executing the computer program, the following steps are further realized:
and tracking the corresponding birds according to the detection result, and performing bird repelling operation.
In an embodiment, after the processor executes the computer program to realize the tracking of the corresponding birds according to the detection result and perform the bird repelling operation step, the processor further realizes the following steps:
and establishing a bird condition database according to the detection result, performing statistical analysis on the bird condition database to obtain an analysis result, and feeding back the analysis result to the terminal.
The storage medium may be a usb disk, a removable hard disk, a Read-Only Memory (ROM), a magnetic disk, or an optical disk, which can store various computer readable storage media.
Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the examples described in connection with the embodiments disclosed herein may be embodied in electronic hardware, computer software, or combinations of both, and that the components and steps of the examples have been described in a functional general in the foregoing description for the purpose of illustrating clearly the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative. For example, the division of each unit is only one logic function division, and there may be another division manner in actual implementation. For example, various elements or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented.
The steps in the method of the embodiment of the invention can be sequentially adjusted, combined and deleted according to actual needs. The units in the device of the embodiment of the invention can be merged, divided and deleted according to actual needs. In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a storage medium. Based on such understanding, the technical solution of the present invention essentially or partially contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a terminal, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention.
While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and various equivalent modifications and substitutions can be easily made by those skilled in the art within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A method of bird detection comprising:
acquiring an image to be detected;
inputting the image to be detected into a bird detection model for detection to obtain a detection result;
feeding back the detection result;
the bird detection model is obtained by adopting an image with a bird label as a sample set to train a deep learning network for the second time.
2. The bird detection method of claim 1, wherein the bird detection model is obtained by secondarily training a deep learning network using an image with a bird tag as a sample set, and comprises:
acquiring an image with a bird tag to obtain a sample set;
constructing a deep learning network;
carrying out coarse precision training on the deep learning network by using a sample set to obtain an intermediate model;
and carrying out high-precision training on the intermediate model to obtain a bird detection model.
3. The bird detection method of claim 2, wherein the sample set is formed by extracting data belonging to birds from a common data set and labeling the bird data photographed by a camera.
4. The bird detection method of claim 2, wherein the training the intermediate model with high precision to obtain a bird detection model comprises:
and training the intermediate model by utilizing bird data shot by the camera in the sample set to obtain a bird detection model.
5. The bird detection method of claim 1, further comprising, after the feeding back the detection result:
and tracking the corresponding birds according to the detection result, and performing bird repelling operation.
6. The bird detection method according to claim 1, wherein after tracking the corresponding birds according to the detection result and performing a bird repelling operation, the method further comprises:
and establishing a bird condition database according to the detection result, performing statistical analysis on the bird condition database to obtain an analysis result, and feeding back the analysis result to the terminal.
7. Bird detection device, its characterized in that includes:
the image acquisition unit is used for acquiring an image to be detected;
the detection unit is used for inputting the image to be detected into a bird detection model for detection so as to obtain a detection result;
and the feedback unit is used for feeding back the detection result.
8. The bird detection device of claim 7, further comprising:
and the model acquisition unit is used for adopting the image with the bird label as a sample set to secondarily train the deep learning network so as to obtain the bird detection model.
9. A computer device, characterized in that the computer device comprises a memory, on which a computer program is stored, and a processor, which when executing the computer program implements the method according to any of claims 1 to 6.
10. A storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011540324.6A CN112633375A (en) | 2020-12-23 | 2020-12-23 | Bird detection method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011540324.6A CN112633375A (en) | 2020-12-23 | 2020-12-23 | Bird detection method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112633375A true CN112633375A (en) | 2021-04-09 |
Family
ID=75322002
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011540324.6A Pending CN112633375A (en) | 2020-12-23 | 2020-12-23 | Bird detection method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112633375A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115500342A (en) * | 2022-09-23 | 2022-12-23 | 国网河北省电力有限公司衡水供电分公司 | Bird repelling device, method, terminal and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110024217A (en) * | 2009-09-01 | 2011-03-09 | 한국전자통신연구원 | Method and apparatus for birds control using mobile robot |
| KR20110066245A (en) * | 2009-12-10 | 2011-06-17 | 한국원자력연구원 | Countermeasure system for birds |
| CN102150653A (en) * | 2011-03-11 | 2011-08-17 | 湖南继善高科技有限公司 | Movable airfield avian detection and directional anti-bird device |
| US20160063310A1 (en) * | 2013-03-28 | 2016-03-03 | Nec Corporation | Bird detection device, bird detection system, bird detection method, and program |
| US20180136650A1 (en) * | 2015-06-29 | 2018-05-17 | Yuneec Technology Co., Limited | Aircraft and obstacle avoidance method and system thereof |
| CN108710126A (en) * | 2018-03-14 | 2018-10-26 | 上海鹰觉科技有限公司 | Automation detection expulsion goal approach and its system |
| CN109122660A (en) * | 2018-08-22 | 2019-01-04 | 深圳威琳懋生物科技有限公司 | Orient bird repellent control method and computer readable storage medium |
| WO2019232830A1 (en) * | 2018-06-06 | 2019-12-12 | 平安科技（深圳）有限公司 | Method and device for detecting foreign object debris at airport, computer apparatus, and storage medium |
| CN111062885A (en) * | 2019-12-09 | 2020-04-24 | 中国科学院自动化研究所 | Mark detection model training and mark detection method based on multi-stage transfer learning |
| CN111709374A (en) * | 2020-06-18 | 2020-09-25 | 深圳市赛为智能股份有限公司 | Bird condition detection method and device, computer equipment and storage medium |
| CN111738354A (en) * | 2020-07-20 | 2020-10-02 | 深圳市天和荣科技有限公司 | Automatic recognition training method, system, storage medium and computer equipment |
-
2020
- 2020-12-23 CN CN202011540324.6A patent/CN112633375A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110024217A (en) * | 2009-09-01 | 2011-03-09 | 한국전자통신연구원 | Method and apparatus for birds control using mobile robot |
| KR20110066245A (en) * | 2009-12-10 | 2011-06-17 | 한국원자력연구원 | Countermeasure system for birds |
| CN102150653A (en) * | 2011-03-11 | 2011-08-17 | 湖南继善高科技有限公司 | Movable airfield avian detection and directional anti-bird device |
| US20160063310A1 (en) * | 2013-03-28 | 2016-03-03 | Nec Corporation | Bird detection device, bird detection system, bird detection method, and program |
| US20180136650A1 (en) * | 2015-06-29 | 2018-05-17 | Yuneec Technology Co., Limited | Aircraft and obstacle avoidance method and system thereof |
| CN108710126A (en) * | 2018-03-14 | 2018-10-26 | 上海鹰觉科技有限公司 | Automation detection expulsion goal approach and its system |
| WO2019232830A1 (en) * | 2018-06-06 | 2019-12-12 | 平安科技（深圳）有限公司 | Method and device for detecting foreign object debris at airport, computer apparatus, and storage medium |
| CN109122660A (en) * | 2018-08-22 | 2019-01-04 | 深圳威琳懋生物科技有限公司 | Orient bird repellent control method and computer readable storage medium |
| CN111062885A (en) * | 2019-12-09 | 2020-04-24 | 中国科学院自动化研究所 | Mark detection model training and mark detection method based on multi-stage transfer learning |
| CN111709374A (en) * | 2020-06-18 | 2020-09-25 | 深圳市赛为智能股份有限公司 | Bird condition detection method and device, computer equipment and storage medium |
| CN111738354A (en) * | 2020-07-20 | 2020-10-02 | 深圳市天和荣科技有限公司 | Automatic recognition training method, system, storage medium and computer equipment |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115500342A (en) * | 2022-09-23 | 2022-12-23 | 国网河北省电力有限公司衡水供电分公司 | Bird repelling device, method, terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111696128B (en) | High-speed multi-target detection tracking and target image optimization method and storage medium | |
| Tack et al. | AnimalFinder: A semi-automated system for animal detection in time-lapse camera trap images | |
| CN111709374B (en) | Bird condition detection method, bird condition detection device, computer equipment and storage medium | |
| CN111709421B (en) | Bird identification method, bird identification device, computer equipment and storage medium | |
| US20140313345A1 (en) | Flying object visual identification system | |
| CN111709372A (en) | Bird repelling method and device, computer equipment and storage medium | |
| CN114677554A (en) | Statistical filtering infrared small target detection tracking method based on YOLOv5 and Deepsort | |
| CN113112480B (en) | Video scene change detection method, storage medium and electronic device | |
| CN112541372B (en) | Difficult sample screening method and device | |
| WO2021157330A1 (en) | Calculator, learning method of discriminator, and analysis system | |
| US11392804B2 (en) | Device and method for generating label objects for the surroundings of a vehicle | |
| JP6787831B2 (en) | Target detection device, detection model generation device, program and method that can be learned by search results | |
| CN113012200B (en) | Method and device for positioning moving object, electronic equipment and storage medium | |
| CN113515968A (en) | Method, device, equipment and medium for detecting street abnormal event | |
| CN112507760A (en) | Method, device and equipment for detecting violent sorting behavior | |
| CN115546260A (en) | Target identification tracking method and device, electronic equipment and storage medium | |
| CN117710756B (en) | Target detection and model training method, device, equipment and medium | |
| CN112633375A (en) | Bird detection method and device, computer equipment and storage medium | |
| CN104077571A (en) | Method for detecting abnormal behavior of throng by adopting single-class serialization model | |
| Delisle et al. | Imperfect detection and wildlife density estimation using aerial surveys with infrared and visible sensors | |
| CN112560621A (en) | Identification method, device, terminal and medium based on animal image | |
| KR102283452B1 (en) | Method and apparatus for disease classification of plant leafs | |
| CN114708645A (en) | Object identification device and object identification method | |
| CN117373108A (en) | Grain storage pest behavior analysis method based on YOLOv5 and improved SORT algorithm | |
| CN115661542A (en) | Small sample target detection method based on feature relation migration |
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 |