CN114170597A - Algae detection equipment and detection method - Google Patents
Algae detection equipment and detection method Download PDFInfo
- Publication number
- CN114170597A CN114170597A CN202111340823.5A CN202111340823A CN114170597A CN 114170597 A CN114170597 A CN 114170597A CN 202111340823 A CN202111340823 A CN 202111340823A CN 114170597 A CN114170597 A CN 114170597A
- Authority
- CN
- China
- Prior art keywords
- algae
- network
- identification model
- yolo
- detection
- 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
- 241000195493 Cryptophyta Species 0.000 title claims abstract description 162
- 238000001514 detection method Methods 0.000 title claims abstract description 65
- 238000012549 training Methods 0.000 claims abstract description 19
- 238000000034 method Methods 0.000 claims abstract description 15
- 238000000605 extraction Methods 0.000 claims description 18
- 230000004927 fusion Effects 0.000 claims description 12
- 238000003384 imaging method Methods 0.000 claims description 11
- 238000004891 communication Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 239000004973 liquid crystal related substance Substances 0.000 claims description 2
- 238000002372 labelling Methods 0.000 claims 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 abstract description 11
- 230000006870 function Effects 0.000 description 17
- 238000010586 diagram Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 229930002875 chlorophyll Natural products 0.000 description 2
- 235000019804 chlorophyll Nutrition 0.000 description 2
- ATNHDLDRLWWWCB-AENOIHSZSA-M chlorophyll a Chemical compound C1([C@@H](C(=O)OC)C(=O)C2=C3C)=C2N2C3=CC(C(CC)=C3C)=[N+]4C3=CC3=C(C=C)C(C)=C5N3[Mg-2]42[N+]2=C1[C@@H](CCC(=O)OC\C=C(/C)CCC[C@H](C)CCC[C@H](C)CCCC(C)C)[C@H](C)C2=C5 ATNHDLDRLWWWCB-AENOIHSZSA-M 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 239000010865 sewage Substances 0.000 description 2
- 231100000331 toxic Toxicity 0.000 description 2
- 230000002588 toxic effect Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 208000003464 asthenopia Diseases 0.000 description 1
- 239000011324 bead Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- DMBHHRLKUKUOEG-UHFFFAOYSA-N diphenylamine Chemical compound C=1C=CC=CC=1NC1=CC=CC=C1 DMBHHRLKUKUOEG-UHFFFAOYSA-N 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
- 238000003911 water pollution 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
- G06F18/21—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
-
- 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
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
- G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
-
- 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
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of algae detection, and discloses algae detection equipment and a detection method, wherein the equipment comprises an embedded platform, a model operating environment is built on the embedded platform, and a trained algae identification model based on a Yolo v4 network is transplanted; training the algae identification model: obtaining a microscopic image of an algae sample, expanding the sample data based on a GAN network, and marking the type and position of algae in the microscopic image; constructing an algae identification model based on a Yolo v4 network, and performing model training on the algae identification model by using marked sample data; when the algae sample is detected, the microscopic image of the algae sample to be detected is imported, detection is carried out through the trained algae identification model, and a detection result is output. The method can quickly and accurately monitor the algae condition in the water body in real time under the condition of insufficient marked samples, is convenient to deploy, has objective and direct prediction results, and greatly reduces the labor cost.
Description
Technical Field
The invention relates to the technical field of algae detection, in particular to algae detection equipment and a detection method.
Background
The research on the distribution of algae in water areas has directional effect on the research on water pollution. Algae are diverse in species and form. In addition, some algae are toxic, and especially toxic algae living in some sewage can bring great influence on the treatment of organic domestic sewage. At present, the traditional method for classifying and identifying algae in a real water sample mainly analyzes the magnitude of the algae from the side by artificial microscopic examination statistics or analyzing the content of chlorophyll in a water body. The manual inspection wastes time and energy, the flow is tedious, the detection result is easily influenced by subjective judgment, and meanwhile, the visual fatigue generated by long-time work also easily causes the reduction of the detection accuracy. The chlorophyll conjecture method cannot accurately obtain the algae distribution, and the cost for obtaining the content of various algae is high.
Disclosure of Invention
The invention aims to solve the technical problem of the prior art and provides algae detection equipment and a detection method.
In order to solve the technical problem, an embodiment of the present invention provides an algae detection apparatus, including a housing and a microscopic imaging apparatus, wherein a circuit board is arranged in the housing, a circuit interface connected to the circuit board is arranged on the housing, the microscopic imaging apparatus is connected to the circuit board through the circuit interface, an embedded platform is arranged on the circuit board, a model operating environment is built on the embedded platform, and a trained algae identification model based on a Yolo v4 network is transplanted;
wherein the training process of the trained algae recognition model based on the Yolo v4 network is as follows: obtaining a microscopic image of an algae sample, expanding the sample data based on a GAN network, and marking the type and position of algae in the microscopic image; constructing an algae identification model based on a Yolo v4 network, and performing model training on the algae identification model by using marked sample data;
when the algae sample is detected, the microscopic imaging equipment is used for acquiring a microscopic image of the algae sample to be detected, detecting and marking the type and the quantity of algae through a trained algae identification model, and outputting a detection result.
The invention has the beneficial effects that: the algae target detection and positioning method can detect and position algae targets, all common algae information can be solidified into the model through training, the algae detection of all water bodies can be conveniently and rapidly realized by using the intelligent detection equipment embedded in the Yolo v4, and the algae density is counted; the method is based on the GAN network and the Yolo v4 network, so that the accuracy of object detection in a complex environment is improved, and the detection precision of the model is higher; the method can effectively learn various algae information, theoretically cover all algae commonly seen in the water body, and the model can permanently memorize the information of the target to be tested through training, thereby greatly reducing the labor cost.
On the basis of the technical scheme, the invention can be further improved as follows.
Further, the Yolo v4 network-based algae recognition model comprises: a feature extraction BackBone network BackBone part, a feature fusion network Neck part and an output network Head part; the feature extraction BackBone network BackBone part adopts a CSPDarknet53 network structure and outputs three feature maps with different sizes; the feature fusion network Neck part adopts an SSP + PAN network structure to fuse feature information of feature graphs with different sizes; the output network Head part divides three branches from the BackBone network backhaul part for feature extraction so as to obtain prediction information of different scales.
Further, the loss function of the Yolo v4 network-based algae recognition model consists of three parts, including: frame regression loss, confidence loss and classification loss based on the CIOU function.
Further, the detecting and marking the type and the quantity of the algae through the trained algae identification model and outputting the detection result comprises: and marking an algae shape area, a prediction label and confidence coefficient in the microscopic image to be detected through the algae identification model, further counting the algae types and the algae quantity, and outputting a final detection result.
Further, still set up power management module, communication function module, GPS big dipper orientation module, touch LCD screen and warning function module on the circuit board.
Furthermore, the embedded platform adopts a Jetson Nano core module.
In order to solve the above technical problems, an embodiment of the present invention provides an algae detection method, including: introducing a microscopic image of an algae sample to be detected, detecting and marking the type and the quantity of algae through the trained algae identification model, and outputting a detection result;
the training process of the algae identification model based on the Yolo v4 network after training is as follows;
obtaining a microscopic image of an algae sample, expanding the sample data based on a GAN network, and marking the type and position of algae in the microscopic image; constructing an algae identification model based on a Yolo v4 network, and performing model training on the algae identification model by using marked sample data; and building a model operation environment in the embedded platform, and transplanting the trained algae identification model to the embedded platform.
On the basis of the technical scheme, the invention can be further improved as follows.
Further, the Yolo v4 network-based algae recognition model comprises: a feature extraction BackBone network BackBone part, a feature fusion network Neck part and an output network Head part; the feature extraction BackBone network BackBone part adopts a CSPDarknet53 network structure and outputs three feature maps with different sizes; the feature fusion network Neck part adopts an SSP + PAN network structure to fuse feature information of feature graphs with different sizes; the output network Head part divides three branches from the BackBone network backhaul part for feature extraction so as to obtain prediction information of different scales.
Further, the loss function of the Yolo v4 network-based algae recognition model consists of three parts, including: frame regression loss, confidence loss and classification loss based on the CIOU function.
Further, the detecting and marking the type and the quantity of the algae through the trained algae identification model and outputting the detection result comprises: and marking an algae shape area, a prediction label and confidence coefficient in the microscopic image to be detected through the algae identification model, further counting the algae types and the algae quantity, and outputting a final detection result.
Additional aspects of the invention and its advantages will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
FIG. 1 is a schematic view of an algae detection apparatus according to an embodiment of the present invention;
FIG. 2 is a basic block diagram of a GAN network;
FIG. 3 is a diagram of an algae identification model based on a Yolo v4 network according to an embodiment of the present invention;
fig. 4 is a flowchart of an algae detection method according to an embodiment of the present invention.
Detailed Description
The embodiments of the present disclosure are described below with specific examples, and other advantages and effects of the present disclosure will be readily apparent to those skilled in the art from the disclosure in the specification. It is to be understood that the described embodiments are merely a subset of the disclosed embodiments and not all embodiments. The disclosure may be embodied or carried out in various other specific embodiments, and various modifications and changes may be made in the details within the description without departing from the spirit of the disclosure. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
It is noted that various aspects of the embodiments are described below within the scope of the appended claims. It should be apparent that the aspects described herein may be embodied in a wide variety of forms and that any specific structure and/or function described herein is merely illustrative. Based on the disclosure, one skilled in the art should appreciate that one aspect described herein may be implemented independently of any other aspects and that two or more of these aspects may be combined in various ways. For example, an apparatus may be implemented and/or a method practiced using any number of the aspects set forth herein. Additionally, such an apparatus may be implemented and/or such a method may be practiced using other structure and/or functionality in addition to one or more of the aspects set forth herein.
Fig. 1 is a schematic view of an algae detection apparatus according to an embodiment of the present invention. As shown in fig. 1, the apparatus includes: casing 10 and micro-imaging device 20, set up circuit board 11 in the casing, set up on the casing with circuit interface 12 that the circuit board is connected, micro-imaging device 20 passes through circuit interface 12 connects circuit board 11, set up embedded platform 111 on the circuit board 11, embedded platform has been built model operating environment, transplants the algae identification model based on Yolo v4 network that has trained to accomplish. The circuit interface 12 includes a video interface, a power interface, a communication interface (a network port, a serial port, a USB interface), and the like.
The embedded platform adopts a Jetson Nano core module, which is a small computer module with strong functions and the size of the small computer module is only 70x45 mm. Multiple neural networks may be run in parallel in applications such as image classification, target detection, segmentation, and speech processing. Jetson Nano can provide 472GFLOPS computational performance for running modern AI algorithms with power consumption of only 5 to 10 watts.
Wherein the training process of the trained algae recognition model based on the Yolo v4 network is as follows: obtaining a microscopic image of an algae sample, expanding the sample data based on a GAN network, and marking the type and position of algae in the microscopic image; an algae identification model based on a Yolo v4 network is constructed, and model training is carried out on the algae identification model by using marked sample data.
The basic structure diagram of GAN is shown in fig. 2, and it is composed of a Generative model (G) and a discriminant model (D). The generative model can be viewed as a sample generator, by inputting a random noise Z and simulating the distribution of real data samples, so that the generated false samples have probability distribution consistent with the real samples as much as possible. The discrimination model is used for discriminating whether the input sample is a real sample or not and outputting the probability that the data is true. G and D compete with each other: g tries to cheat D so as to falsely and truly, D continuously improves the discrimination capability to prevent the data synthesized by G from being mixed with the beads, and theoretically, the generated data distribution Pg is infinitely close to the real data distribution Pdata. The GAN network ensemble optimization function is shown below.
When the algae sample is detected, the microscopic imaging equipment is used for acquiring a microscopic image of the algae sample to be detected, detecting and marking the type and the quantity of algae through a trained algae identification model, and outputting a detection result.
The algae detection equipment provided by the embodiment can detect and position algae targets, can solidify all common algae information into a model through training, can conveniently and quickly realize algae detection of all water bodies by using the intelligent detection equipment embedded with Yolo v4, and can count the algae density; according to the method, the algae identification model of the Yolo v4 network is adopted, and various learning strategies and skills are combined, so that the accuracy of object detection in a complex environment can be improved, and the detection precision of the model is higher; the method can effectively learn various algae information, theoretically cover all the algae commonly seen in the water body, and the model can permanently memorize the information of the target to be tested through training. The embodiment of the invention can quickly and accurately monitor the algae condition in the water body in real time under the condition of insufficient marked samples, is convenient to deploy, has objective and direct prediction results, and greatly reduces the labor cost.
Optionally, as shown in fig. 3, the Yolo v4 network-based algae recognition model includes: a feature extraction BackBone network BackBone part, a feature fusion network Neck part and an output network Head part; the feature extraction BackBone network BackBone part adopts a CSPDarknet53 network structure and outputs three feature maps with different sizes; the feature fusion network hack part adopts an SSP (spatial Pyramid Pooling) + PAN network structure to fuse feature information of feature maps with different sizes; the Head part of the output network divides three branches from the feature extraction backbone network to obtain prediction information with different scales.
The backbone network comprises three output branches, wherein one output branch is SSP output, and the SSP network structure is adopted to fuse the feature information of feature graphs with different sizes; the other two branches of the backbone network and the features extracted by the SPP branch are synchronously input into a PANet structure of a Neck part of the network, different prediction branches are constructed, and the comprehensive prediction of algae information is realized.
Optionally, the loss function of the Yolo v4 network-based algae recognition model consists of three parts, including: frame regression loss, confidence loss and classification loss based on the CIOU function.
Optionally, the detecting and marking the type and quantity of the algae through the trained algae recognition model, and outputting a detection result, including: and marking an algae shape area, a prediction label and confidence coefficient in the microscopic image to be detected through the algae identification model, giving a probability value of the class to which the detection target belongs, further counting the algae type and quantity, and outputting a final detection result.
The circuit board is also provided with a power management module 112, a communication function module 113, a GPS/Beidou positioning module 114, a touch liquid crystal screen 115, a warning function module 116 and the like.
The communication function module 113 mainly includes a video interface for connecting with the microscopic imaging device and a network communication interface for transmitting the field data to the server. The GPS/beidou positioning module 114 is configured to send the positioning information to the server, so that the position of the water body can be determined through the coordinate information. The touch screen 115 facilitates field debugging, such as modifying device parameters and the like (sampling intervals, built-in model parameters and the like). The alert function 116 primarily implements indications of equipment failures, or indications in the case of significant over-limits, etc.
As shown in fig. 2, an embodiment of the present invention provides an algae detection method, including the following steps:
s210, obtaining a microscopic image of the algae sample, expanding the sample data based on a GAN network, and marking the type and position of algae in the microscopic image.
Specifically, the microscopic image can be labeled by means of a LabelImg.exe tool, and the shape regions and names of different algae are labeled to generate a ". xml" file; sample labels are modified prior to training.
S220, constructing an algae identification model based on a Yolo v4 network, and performing model training on the algae identification model by using marked sample data.
The algae identification model based on the Yolo v4 network comprises: a feature extraction BackBone network BackBone part, a feature fusion network Neck part and an output network Head part; the feature extraction BackBone network BackBone part adopts a CSPDarknet53 network structure and outputs three feature maps with different sizes; the feature fusion network Neck part adopts an SSP + PAN network structure to fuse feature information of feature graphs with different sizes; the output network Head part divides three branches from the BackBone network BackBone part for feature extraction so as to obtain prediction information of different scales.
The loss function of the algae recognition model based on the Yolo v4 network consists of three parts, including: frame regression loss, confidence loss and classification loss based on the CIOU function.
And S230, building a model operating environment on the embedded switch platform, and transplanting the trained algae identification model to the embedded platform. The embedded platform can adopt a Jetson Nano core module, and can retrain the algae identification model deployed in a Jetson Nano suite.
S240, importing the microscopic image of the algae sample to be detected, detecting and marking the type and the quantity of the algae through the trained algae identification model, and outputting the detection result.
Specifically, during sample detection, the water to be detected can be drained to the sampling plate or the sampling pool 30, and a microscopic image of the algae sample to be detected is obtained through the microscopic imaging device. And marking an algae shape area, a prediction label and confidence coefficient in the microscopic image to be detected through the trained algae identification model, giving a probability value of the class to which the detection target belongs, further counting the algae type and quantity, and outputting a final detection result. And storing the detection result in real time and uploading the detection result to a background server. Manual verification of the test structure is also possible.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. The algae detection equipment is characterized by comprising a shell and microscopic imaging equipment, wherein a circuit board is arranged in the shell, a circuit interface connected with the circuit board is arranged on the shell, the microscopic imaging equipment is connected with the circuit board through the circuit interface, an embedded platform is arranged on the circuit board, a model operating environment is built on the embedded platform, and a trained algae identification model based on a Yolo v4 network is transplanted;
wherein the training process of the trained Yolo v4 network-based algae recognition model comprises the following steps: obtaining a microscopic image of an algae sample, expanding the sample data based on a GAN network, and marking the type and position of algae in the microscopic image; constructing an algae identification model based on a Yolo v4 network, and performing model training on the algae identification model by using marked sample data;
when the algae sample is detected, the microscopic imaging equipment is used for acquiring a microscopic image of the algae sample to be detected, detecting and marking the type and the quantity of algae through a trained algae identification model, and outputting a detection result.
2. The algae detection apparatus of claim 1, wherein the Yolo v4 network-based algae recognition model comprises: a feature extraction BackBone network BackBone part, a feature fusion network Neck part and an output network Head part;
the feature extraction BackBone network BackBone part adopts a CSPDarknet53 network structure and outputs three feature maps with different sizes; the feature fusion network Neck part adopts an SSP + PAN network structure to fuse feature information of feature graphs with different sizes; the output network Head part divides three branches from the BackBone network BackBone part for feature extraction so as to obtain prediction information of different scales.
3. The algae detection apparatus of claim 2, wherein the loss function of the Yolo v4 network-based algae recognition model consists of three parts including: frame regression loss, confidence loss and classification loss based on the CIOU function.
4. The algae detection apparatus of any one of claims 1 to 3, wherein the detecting and marking the type and amount of algae by the trained algae recognition model and outputting the detection result comprises: and marking an algae shape area, a prediction label and confidence coefficient in the microscopic image to be detected through the algae identification model, further counting the algae types and the algae quantity, and outputting a final detection result.
5. The algae detection apparatus of any one of claims 1 to 3, wherein the circuit board further integrates a power management module, a communication function module, a GPS/Beidou positioning module, a touch liquid crystal screen and a warning function module.
6. The algae detection apparatus of any one of claims 1 to 3, wherein the embedded platform employs a Jetson Nano core module.
7. An algae detection method, comprising:
introducing a microscopic image of an algae sample to be detected, detecting and marking the type and the quantity of algae through the trained algae identification model, and outputting a detection result;
wherein the training process of the trained algae recognition model based on the Yolo v4 network comprises the following steps;
obtaining a microscopic image of an algae sample, expanding the sample data based on a GAN network, and marking the type and position of algae in the microscopic image;
constructing an algae identification model based on a Yolo v4 network, and performing model training on the algae identification model by using marked sample data;
and building a model operation environment in the embedded platform, and transplanting the trained algae identification model to the embedded platform.
8. The algae detection method of claim 7, wherein the Yolo v4 network-based algae recognition model comprises: a feature extraction BackBone network BackBone part, a feature fusion network Neck part and an output network Head part;
the feature extraction BackBone network BackBone part adopts a CSPDarknet53 network structure and outputs three feature maps with different sizes; the feature fusion network Neck part adopts an SSP + PAN network structure to fuse feature information of feature graphs with different sizes; the output network Head part divides three branches from the BackBone network BackBone part for feature extraction so as to obtain prediction information of different scales.
9. The algae detection method of claim 7, wherein the loss function of the Yolo v4 network-based algae recognition model consists of three parts including: frame regression loss, confidence loss and classification loss based on the CIOU function.
10. The algae detection method of any one of claims 7 to 9, wherein the detecting and labeling the algae species and quantity through the trained algae recognition model and outputting the detection result comprises: and marking an algae shape area, a prediction label and confidence coefficient in the microscopic image to be detected through the algae identification model, further counting the algae types and the algae quantity, and outputting a final detection result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111340823.5A CN114170597A (en) | 2021-11-12 | 2021-11-12 | Algae detection equipment and detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111340823.5A CN114170597A (en) | 2021-11-12 | 2021-11-12 | Algae detection equipment and detection method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114170597A true CN114170597A (en) | 2022-03-11 |
Family
ID=80478817
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111340823.5A Pending CN114170597A (en) | 2021-11-12 | 2021-11-12 | Algae detection equipment and detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114170597A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115187982A (en) * | 2022-07-12 | 2022-10-14 | 河北华清环境科技集团股份有限公司 | Algae detection method and device and terminal equipment |
CN116797601A (en) * | 2023-08-24 | 2023-09-22 | 西南林业大学 | Image recognition-based Huashansong growth dynamic monitoring method and system |
CN117115573A (en) * | 2023-10-25 | 2023-11-24 | 华侨大学 | Toxic biological image classification and identification method, device, equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111582345A (en) * | 2020-04-29 | 2020-08-25 | 中国科学院重庆绿色智能技术研究院 | Target identification method for complex environment under small sample |
CN112801164A (en) * | 2021-01-22 | 2021-05-14 | 北京百度网讯科技有限公司 | Training method, device and equipment of target detection model and storage medium |
CN113159024A (en) * | 2021-03-22 | 2021-07-23 | 河海大学 | License plate recognition technology based on improved YOLOv4 |
-
2021
- 2021-11-12 CN CN202111340823.5A patent/CN114170597A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111582345A (en) * | 2020-04-29 | 2020-08-25 | 中国科学院重庆绿色智能技术研究院 | Target identification method for complex environment under small sample |
CN112801164A (en) * | 2021-01-22 | 2021-05-14 | 北京百度网讯科技有限公司 | Training method, device and equipment of target detection model and storage medium |
CN113159024A (en) * | 2021-03-22 | 2021-07-23 | 河海大学 | License plate recognition technology based on improved YOLOv4 |
Non-Patent Citations (1)
Title |
---|
赵宇航等: "基于YOLO V3的无人机航拍车辆检测方法", 《电子世界》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115187982A (en) * | 2022-07-12 | 2022-10-14 | 河北华清环境科技集团股份有限公司 | Algae detection method and device and terminal equipment |
CN116797601A (en) * | 2023-08-24 | 2023-09-22 | 西南林业大学 | Image recognition-based Huashansong growth dynamic monitoring method and system |
CN116797601B (en) * | 2023-08-24 | 2023-11-07 | 西南林业大学 | Image recognition-based Huashansong growth dynamic monitoring method and system |
CN117115573A (en) * | 2023-10-25 | 2023-11-24 | 华侨大学 | Toxic biological image classification and identification method, device, equipment and storage medium |
CN117115573B (en) * | 2023-10-25 | 2024-01-05 | 华侨大学 | Toxic biological image classification and identification method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114170597A (en) | Algae detection equipment and detection method | |
CN103528617B (en) | A kind of cockpit instrument identifies and detection method and device automatically | |
CN108305243A (en) | A kind of magnetic tile surface defect detection method based on deep learning | |
CN108509976A (en) | The identification device and method of animal | |
CN110222701A (en) | A kind of bridge defect automatic identifying method | |
CN107229930A (en) | A kind of pointer instrument numerical value intelligent identification Method and device | |
Leow et al. | Automated identification of copepods using digital image processing and artificial neural network | |
CN1806501B (en) | Marine phytoplankton automatic distinguishing method and apparatus | |
CN110135271A (en) | A kind of cell sorting method and device | |
CN107808358A (en) | Image watermark automatic testing method | |
CN108090906A (en) | A kind of uterine neck image processing method and device based on region nomination | |
CN116359218B (en) | Industrial aggregation area atmospheric pollution mobile monitoring system | |
CN112613454A (en) | Electric power infrastructure construction site violation identification method and system | |
CN116363440B (en) | Deep learning-based identification and detection method and system for colored microplastic in soil | |
CN114494837A (en) | Intelligent density identification method and system for fishery resources | |
CN113128385A (en) | Toxic algae monitoring and early warning method and system | |
CN112036384B (en) | Sperm head shape recognition method, device and equipment | |
CN113128335A (en) | Method, system and application for detecting, classifying and discovering micro-body paleontological fossil image | |
CN109492534A (en) | A kind of pedestrian detection method across scene multi-pose based on Faster RCNN | |
CN112200011A (en) | Aeration tank state detection method and system, electronic equipment and storage medium | |
CN110175519A (en) | Method and device for identifying separation and combination identification instrument of transformer substation and storage medium | |
CN114387235B (en) | Water environment monitoring method and system | |
CN117115722A (en) | Construction scene detection method and device, storage medium and electronic equipment | |
CN114005064A (en) | Biological water body pollution early warning method and device based on machine vision technology | |
CN115311657B (en) | Multi-source algae image target detection method, system, electronic equipment and storage medium |
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: 20220311 |
|
RJ01 | Rejection of invention patent application after publication |