CN115690566A - Deep sea animal new species identification method based on deep migration clustering learning - Google Patents
Deep sea animal new species identification method based on deep migration clustering learning Download PDFInfo
- Publication number
- CN115690566A CN115690566A CN202211299614.5A CN202211299614A CN115690566A CN 115690566 A CN115690566 A CN 115690566A CN 202211299614 A CN202211299614 A CN 202211299614A CN 115690566 A CN115690566 A CN 115690566A
- Authority
- CN
- China
- Prior art keywords
- deep
- classification network
- class classification
- picture
- new species
- 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.)
- Granted
Links
- 241000894007 species Species 0.000 title claims abstract description 67
- 241001465754 Metazoa Species 0.000 title claims abstract description 66
- 230000005012 migration Effects 0.000 title claims abstract description 47
- 238000013508 migration Methods 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 239000011159 matrix material Substances 0.000 claims description 42
- 230000009467 reduction Effects 0.000 claims description 17
- 238000000513 principal component analysis Methods 0.000 claims description 6
- 238000003064 k means clustering Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 3
- 230000001737 promoting effect Effects 0.000 claims description 2
- 238000012847 principal component analysis method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 239000004744 fabric Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- 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
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/80—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in fisheries management
- Y02A40/81—Aquaculture, e.g. of fish
Abstract
The invention discloses a deep sea animal new species identification method and a system based on deep migration clustering learning, wherein the method comprises the following steps: taking a deep sea animal picture and preprocessing the deep sea animal picture to obtain a preprocessed picture; constructing a known class classification network and an unknown class classification network, training the networks and learning new species characteristics to obtain a deep migration clustering learning model; and performing species identification on the preprocessed pictures based on a deep migration clustering learning model. The system comprises: the device comprises a shooting module, a model building module and an identification module. By using the method, species classification annotation and identification of new species of potential deep sea animals can be carried out on the discovered deep sea animals. The method and the system for identifying the new species of the deep sea animals based on the deep migration clustering learning can be widely applied to the field of image identification.
Description
Technical Field
The invention relates to the field of image recognition, in particular to a deep-sea animal new species recognition method based on deep migration clustering learning.
Background
As world countries continuously advance deep sea exploration, the information base of world marine animal species is huge and is increased day by day, and when the unmanned remotely operated vehicle works in deep sea, the deep sea animal species information is difficult to be found and collected in a targeted manner in the deep sea in a short time. More in the traditional identification of the new species of the deep sea animals, experts judge according to self knowledge storage and network information retrieval, a large amount of human resources and time cost are needed to be paid by the method, subjective judgment depending on the experts is needed, if the experts cannot make effective judgment in time, the deep sea exploration research progress is slow or the discovery of the new species of the deep sea animals is missed, and then the identification is carried out by manpower finally, so that the continuity of the research work needs to be improved.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a deep sea animal new species identification method based on deep migration clustering learning, which can reduce the dependency of the discovery of the deep sea animal new species on experts, does not need to manually set any confidence domain value to give an objective and effective judgment result, and effectively learns the characteristics of the new species in the process of completing the discovery of the new species.
The first technical scheme adopted by the invention is as follows: a deep-sea animal new species identification method based on deep migration clustering learning comprises the following steps:
taking a deep sea animal picture and preprocessing the deep sea animal picture to obtain a preprocessed picture;
constructing a known class classification network and an unknown class classification network, training the networks and learning new species characteristics to obtain a deep migration clustering learning model;
and performing species identification on the preprocessed pictures based on a deep migration clustering learning model.
Further, the step of taking a deep sea animal picture and preprocessing the deep sea animal picture to obtain a preprocessed picture specifically includes:
shooting a deep sea animal picture based on the unmanned remote control submersible;
cutting the deep-sea animal photo according to a preset size to obtain a cut picture;
and performing data enhancement on the cut picture to obtain a preprocessed picture.
Further, the step of constructing a known class classification network and an unknown class classification network, training the networks and learning new species characteristics to obtain a deep migration clustering learning model specifically includes:
constructing a known class classification network based on the VGG16 and training to obtain the trained known class classification network;
constructing an unknown class classification network based on VGG16 and training, promoting the network to autonomously learn the characteristics of a new species, and obtaining the trained unknown class classification network;
and integrating the trained known class classification network and the trained unknown class classification network to obtain the deep migration clustering learning model.
Further, the training steps of the known class classification network are as follows:
acquiring a training data set with class marks and inputting the training data set to a known class classification network;
outputting a first encoding vector matrix based on a feature extractor of a known class classification network;
reducing the dimension of the first coding vector matrix based on the principal component analysis dimension reduction layer to obtain a first coding vector matrix after dimension reduction;
clustering the first code vector matrix subjected to dimensionality reduction based on a K-means clustering algorithm to obtain a first clustering center set;
performing one-hot coding on the class labels in the training data set to obtain a target distribution matrix;
and (3) taking the cross entropy loss of the probability distribution matrix and the target distribution matrix of the minimized sample distributed to the first clustering center set as a target, and adjusting the parameters of the known class classification network (the adjusted network parameters of the feature extractor and the parameters of the PCA principal component analysis dimensionality reduction layer, namely the kmean clustering center set) through the random gradient descent back propagation to obtain the trained known class classification network.
Further, the expression of the probability distribution matrix of the sample assigned to the first cluster center set is as follows:
P＝(1+||z a -U b || 2 ) -1
in the above equation, P denotes that each sample is assigned to U b In the probability distribution matrix, U, of each cluster center b Representing a set of cluster centers, z a And representing the code vector matrix after dimension reduction.
Further, the training step of the unknown class network is as follows:
constructing a target data set based on the preprocessed pictures and inputting the target data set to an unknown class network;
performing feature extraction on the target data set by using a feature extractor of a known class classification network to obtain a second coding vector matrix;
reducing the dimension of the second coding vector matrix based on the principal component analysis dimension reduction layer to obtain a second coding vector matrix after dimension reduction;
clustering the second code vector matrix subjected to dimensionality reduction based on a K-means clustering algorithm to obtain a second clustering center set;
constructing a target probability distribution matrix based on a Bayesian criterion;
and (3) taking the probability distribution matrix and the target distribution matrix cross entropy loss of the minimized sample distributed to the second clustering center set as targets, and adjusting the parameters of the unknown class classification network (the adjusted network parameters of the feature extractor and the parameters of the PCA principal component analysis dimensionality reduction layer, namely the kmean clustering center set) through the random gradient descent back propagation to obtain the trained unknown class classification network.
Further, the step of performing species identification on the preprocessed picture based on the deep migration clustering learning model specifically includes:
selecting a target picture from the preprocessed pictures and inputting the target picture into a known class classification network and an unknown class classification network in the deep migration clustering learning model respectively;
outputting a first classification label based on the unknown class classification network;
outputting a second classification label based on the known class classification network;
retrieving the deep sea animal photo corresponding to the second classification label in the training data set to obtain a retrieval picture;
inputting the retrieval picture into an unknown class classification network to obtain a retrieval label;
judging that the retrieval label is the same as the first classification label, and determining that the deep-sea animal in the target picture does not belong to a new species;
and judging that the retrieval label is not the same as the first classification label, and identifying that the deep sea animal in the target picture belongs to a new species.
The second technical scheme adopted by the invention is as follows: a deep-sea animal new species identification system based on deep migration clustering learning comprises:
the shooting module is used for shooting a deep sea animal picture and carrying out pretreatment to obtain a pretreated picture;
the model building module is used for building a known class classification network and an unknown class classification network, training the networks and learning new species characteristics to obtain a deep migration clustering learning model;
and the identification module is used for carrying out species identification on the preprocessed pictures based on the deep migration clustering learning model.
The method, the system, the device and the storage medium have the advantages that: according to the invention, the deep sea animal photo shot by the unmanned remote-control submersible is obtained, the deep sea animal photo is processed, and the deep sea animal species data set pre-trained deep migration clustering learning model is used for species identification, so that species classification and annotation of the found deep sea animal and identification of a new species of a potential deep sea animal are realized in a short time, technical personnel in the field are not required to judge whether the species is a new species according to the domain value setting of confidence, the dependency of the new species discovery of the deep sea animal on experts is reduced, and the discovered new species characteristics can be learned while the species are classified.
Drawings
FIG. 1 is a flow chart of the steps of a deep sea animal new species identification method based on deep migration clustering learning of the invention;
FIG. 2 is a diagram illustrating an exemplary application scenario of an embodiment of the present invention;
FIG. 3 is a schematic diagram of a VGG16 architecture with the classifier layer removed in accordance with an embodiment of the present invention;
FIG. 4 is a schematic diagram of a known class classification network training process according to an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating a training process of an unknown class classification network according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a general flow chart of species identification according to an embodiment of the present invention.
FIG. 7 is a structural block diagram of the deep sea animal new species identification system based on deep migration clustering learning.
Detailed Description
The invention is described in further detail below with reference to the figures and the specific embodiments. For the step numbers in the following embodiments, they are set for convenience of illustration only, the order between the steps is not limited at all, and the execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.
As shown in figure 1, the invention provides a deep sea animal new species identification method based on deep migration clustering learning, which comprises the following steps:
s1, shooting a deep sea animal picture and preprocessing the deep sea animal picture to obtain a preprocessed picture;
s1.1, shooting a deep sea animal picture based on an unmanned remote control submersible vehicle;
specifically, referring to fig. 2, the unmanned remotely operated vehicle is controlled by an operator at a work station at sea in real time, and the operator remotely controls the unmanned remotely operated vehicle through a computer terminal to shoot and identify i target pictures to form a target image set x i And transmitting to the offshore workstation server in real time.
S1.2, cutting the deep sea animal photo according to a preset size to obtain a cut picture;
specifically, the target image set x i The length and width of the cloth are cut into 224 and 244 respectively.
And S1.3, performing data enhancement on the cut picture to obtain a preprocessed picture.
Specifically, the target image set x is ensured i For the target image set x under the condition that the size of the picture is not changed i Randomly changing (such as turning, stretching, cutting, deforming) the target image set x i Each picture can obtain j enhanced pictures after data enhancement, and a target picture set x with the number of pictures r (r = i + i × j) is formed r 。
S2, constructing a known class classification network and an unknown class classification network, training the networks and learning new species characteristics to obtain a deep migration clustering learning model;
s2.1, a known class classification network (the VGG16 network without the classifier layer, the PCA dimension reduction layer and the kmeans clustering layer form a classification network): VGG16 (visual geometry group network architecture 16, see FIG. three) with removed classifier layer is used as a feature extractor for the known classification network omega
The data set C of the deep sea animals in the known category with the length and width of the preprocessed picture being 224 x 224, the number of samples being a and the number of the known categories being b a As an input to the VGG16, such as a fisherbase dataset,
reducing the output code vector matrix M to b dimension by using PCA (principal component analysis method) to obtain a code vector matrix z a By pair z a Clustering by K-means (K mean value clustering algorithm) to obtain a clustering center set U b . Each sample is assigned to U b The probability distribution matrix P for each cluster center in (a) is defined as follows:
P＝(1+||z a -U b || 2 ) -1
c is to be a The label is subjected to one-hot coding (one-hot coding) to obtain a target distribution matrix q, cross entropy loss of P and q is minimized, parameters of a model are adjusted through random gradient descent and back propagation, and a known classification network omega pre-training process is shown in figure 4.
S2.2, unknown class classification network: use of
For x r Carrying out feature extraction to obtain a coding vector matrix N r To N, to r Using PCA (principal component analysis method) to reduce to b +1 dimension to obtain coding vector matrix Y r By the pair Y r Performing K-means clustering to obtain a clustering center set O b+1 Constructing a probability distribution matrix p' with individual samples assigned to O b+1 The probability p' (r | k) of each cluster center in (1) is defined as follows:
p′ (r|k) ＝(1+||Y r -O b+1 || 2 ) -1
based on Bayesian criterion, a target probability distribution matrix q' is constructed, and single samples are distributed to O b+1 The target probability q' (r | k) of each cluster center in (1) is defined as follows:
the training flow of the unknown class learning classification network ρ is shown in fig. 5.
And S3, performing species identification on the preprocessed pictures based on the deep migration clustering learning model.
Specifically, after ρ learning is completed, from x i Selecting a target picture x as input of rho to obtain a classification label Y, then using x as input of omega to obtain a classification label Y, retrieving a deep sea species picture F corresponding to the label Y in a training data set, inputting rho into F to obtain a label Y ', if the label Y' is the same as the Y, determining that the current target is not a new species, and marking the label Y on the current target; if the labels Y' and Y are not the same, then the situation that the current target and the animal species with the highest possibility of being classified have a decisive characteristic which causes the two to be distinguished is shown, and the deep sea animal belongs to the new species. Finally, species recognition is carried out by using a deep migration clustering learning model pre-trained by a deep sea animal species data set, which is shown in figure 6.
As shown in fig. 7, a deep sea animal new species identification system based on deep migration clustering learning comprises:
the shooting module is used for shooting a deep sea animal picture and carrying out pretreatment to obtain a pretreated picture;
the model building module is used for building a known class classification network and an unknown class classification network, training the networks and learning new species characteristics to obtain a deep migration clustering learning model;
and the identification module is used for carrying out species identification on the preprocessed pictures based on the deep migration clustering learning model.
The contents in the above method embodiments are all applicable to the present system embodiment, the functions specifically implemented by the present system embodiment are the same as those in the above method embodiment, and the beneficial effects achieved by the present system embodiment are also the same as those achieved by the above method embodiment.
A deep sea animal new species recognition device based on deep migration clustering learning:
at least one processor;
at least one memory for storing at least one program;
when the at least one program is executed by the at least one processor, the at least one processor implements the method for identifying new species of abyssal animals based on deep migration clustering learning as described above.
The contents in the method embodiments are all applicable to the device embodiments, the functions specifically implemented by the device embodiments are the same as those in the method embodiments, and the beneficial effects achieved by the device embodiments are also the same as those achieved by the method embodiments.
A storage medium having stored therein instructions executable by a processor, the storage medium comprising: the processor-executable instructions, when executed by the processor, are for implementing a deep sea animal new species identification method based on deep migration cluster learning as described above.
The contents in the foregoing method embodiments are all applicable to this storage medium embodiment, the functions specifically implemented by this storage medium embodiment are the same as those in the foregoing method embodiments, and the beneficial effects achieved by this storage medium embodiment are also the same as those achieved by the foregoing method embodiments.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. A deep sea animal new species identification method based on deep migration clustering learning is characterized by comprising the following steps:
taking a deep sea animal picture and preprocessing the deep sea animal picture to obtain a preprocessed picture;
constructing a known class classification network and an unknown class classification network, training the networks and learning new species characteristics to obtain a deep migration clustering learning model;
and performing species identification on the preprocessed pictures based on a deep migration clustering learning model.
2. The method for identifying the new species of the deep sea animals based on the deep migration clustering learning of claim 1, wherein the step of taking the picture of the deep sea animals and preprocessing the picture to obtain the preprocessed picture comprises the following specific steps:
shooting a deep sea animal picture based on the unmanned remote control submersible;
cutting the deep-sea animal photo according to a preset size to obtain a cut picture;
and performing data enhancement on the cut picture to obtain a preprocessed picture.
3. The deep sea animal new species identification method based on deep migration clustering learning of claim 2, wherein the step of constructing a known class classification network and an unknown class classification network, training the networks and learning new species features to obtain a deep migration clustering learning model specifically comprises:
constructing a known class classification network based on the VGG16 and training to obtain a trained known class classification network;
constructing an unknown class classification network based on VGG16 and training, promoting the network to autonomously learn the characteristics of a new species, and obtaining the trained unknown class classification network;
and integrating the trained known class classification network and the trained unknown class classification network to obtain the deep migration clustering learning model.
4. The deep sea animal new species identification method based on deep migration clustering learning as claimed in claim 3, wherein the training steps of the known class classification network are as follows:
acquiring a training data set with a class label and inputting the training data set to a known class classification network;
outputting a first encoding vector matrix based on a feature extractor of a known class classification network;
reducing the dimension of the first coding vector matrix based on the principal component analysis dimension reduction layer to obtain a first coding vector matrix after dimension reduction;
clustering the first code vector matrix subjected to dimensionality reduction based on a K-means clustering algorithm to obtain a first clustering center set;
carrying out one-hot encoding on the class labels in the training data set to obtain a target distribution matrix;
and taking the cross entropy loss of the probability distribution matrix and the target distribution matrix of the minimized samples distributed to the first clustering center set as a target, and adjusting the parameters of the known class classification network through the random gradient descent back propagation to obtain the trained known class classification network.
5. The method for identifying new species in deep sea animals based on deep migration clustering learning as claimed in claim 4, wherein the expression of the probability distribution matrix of the samples to be assigned to the first cluster center set is as follows:
P＝(1+||z a -U b || 2 ) -1
in the above equation, P denotes that each sample is assigned to U b In the probability distribution matrix, U, of each cluster center b Representing a set of cluster centers, z a And representing the code vector matrix after dimension reduction.
6. The deep sea animal new species identification method based on deep migration clustering learning as claimed in claim 5, characterized in that the training step of the unknown class network is as follows:
constructing a target data set based on the preprocessed pictures and inputting the target data set to an unknown category network;
performing feature extraction on the target data set by using a feature extractor of a known class classification network to obtain a second coding vector matrix;
reducing the dimension of the second coding vector matrix based on the principal component analysis dimension reduction layer to obtain a second coding vector matrix after dimension reduction;
clustering the second code vector matrix subjected to dimensionality reduction based on a K-means clustering algorithm to obtain a second clustering center set;
constructing a target probability distribution matrix based on a Bayesian criterion;
and taking the cross entropy loss of the probability distribution matrix and the target distribution matrix of the minimized samples distributed to the second clustering center set as a target, and adjusting the parameters of the unknown class classification network through the random gradient descent back propagation to obtain the trained unknown class classification network.
7. The deep sea animal new species identification method based on deep migration cluster learning of claim 6, wherein the deep migration cluster learning model-based species identification step of the preprocessed pictures specifically comprises:
selecting a target picture from the preprocessed pictures and inputting the target picture into a known class classification network and an unknown class classification network in the deep migration clustering learning model respectively;
outputting a first classification label based on the unknown class classification network;
outputting a second classification label based on the known class classification network;
retrieving the deep sea animal photo corresponding to the second classification label in the training data set to obtain a retrieval picture;
inputting the retrieval picture into an unknown category classification network to obtain a retrieval label;
judging that the retrieval label is the same as the first classification label, and determining that the deep-sea animal in the target picture does not belong to a new species;
and judging that the retrieval label is different from the first classification label, and determining that the deep-sea animal in the target picture belongs to a new species.
8. A deep sea animal new species identification system based on deep migration clustering learning is characterized by comprising the following steps:
the shooting module is used for shooting a deep sea animal picture and carrying out pretreatment to obtain a pretreated picture;
the model building module is used for building a known class classification network and an unknown class classification network, training the networks and learning new species characteristics to obtain a deep migration clustering learning model;
and the identification module is used for identifying species of the preprocessed pictures based on the deep migration clustering learning model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211299614.5A CN115690566B (en) | 2022-10-24 | 2022-10-24 | Deep sea animal new species identification method based on deep migration cluster learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211299614.5A CN115690566B (en) | 2022-10-24 | 2022-10-24 | Deep sea animal new species identification method based on deep migration cluster learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115690566A true CN115690566A (en) | 2023-02-03 |
| CN115690566B CN115690566B (en) | 2024-01-19 |
Family
ID=85066652
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211299614.5A Active CN115690566B (en) | 2022-10-24 | 2022-10-24 | Deep sea animal new species identification method based on deep migration cluster learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115690566B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116468984A (en) * | 2023-03-10 | 2023-07-21 | 衡阳师范学院 | Construction method of movable object detection model, detection model and detection method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110580496A (en) * | 2019-07-11 | 2019-12-17 | 南京邮电大学 | Deep migration learning system and method based on entropy minimization |
| WO2021022816A1 (en) * | 2019-08-07 | 2021-02-11 | 南京硅基智能科技有限公司 | Intent identification method based on deep learning network |
| CN113283514A (en) * | 2021-05-31 | 2021-08-20 | 高新兴科技集团股份有限公司 | Unknown class classification method, device and medium based on deep learning |
| CN113326289A (en) * | 2021-08-02 | 2021-08-31 | 山东大学 | Rapid cross-modal retrieval method and system for incremental data carrying new categories |
| CN113705507A (en) * | 2021-09-02 | 2021-11-26 | 上海交通大学 | Mixed reality open set human body posture recognition method based on deep learning |
| CN113705580A (en) * | 2021-08-31 | 2021-11-26 | 西安电子科技大学 | Hyperspectral image classification method based on deep migration learning |
-
2022
- 2022-10-24 CN CN202211299614.5A patent/CN115690566B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110580496A (en) * | 2019-07-11 | 2019-12-17 | 南京邮电大学 | Deep migration learning system and method based on entropy minimization |
| WO2021022816A1 (en) * | 2019-08-07 | 2021-02-11 | 南京硅基智能科技有限公司 | Intent identification method based on deep learning network |
| CN113283514A (en) * | 2021-05-31 | 2021-08-20 | 高新兴科技集团股份有限公司 | Unknown class classification method, device and medium based on deep learning |
| CN113326289A (en) * | 2021-08-02 | 2021-08-31 | 山东大学 | Rapid cross-modal retrieval method and system for incremental data carrying new categories |
| CN113705580A (en) * | 2021-08-31 | 2021-11-26 | 西安电子科技大学 | Hyperspectral image classification method based on deep migration learning |
| CN113705507A (en) * | 2021-09-02 | 2021-11-26 | 上海交通大学 | Mixed reality open set human body posture recognition method based on deep learning |
Non-Patent Citations (4)
| Title |
|---|
| KAI HAN ET.AL: "Learning to Discover Novel Visual Categories via Deep Transfer Clustering", 《2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)》, pages 8400 - 8408 * |
| MAGGU, J ET.AL: "Deep Convolutional Transform Learning - Extended version", 《ARXIV》, pages 1 - 10 * |
| 朱顺德;徐增兴;刘一?;: "基于CNN迁移学习的示功图图形分类预警", 中国管理信息化, no. 09, pages 155 - 157 * |
| 王胜: "基于深度学习的属性网络表示学习方法研究", 《中国优秀硕士学位论文全文数据库 基础科学辑》, pages 002 - 36 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116468984A (en) * | 2023-03-10 | 2023-07-21 | 衡阳师范学院 | Construction method of movable object detection model, detection model and detection method |
| CN116468984B (en) * | 2023-03-10 | 2023-10-27 | 衡阳师范学院 | Construction method of movable object detection model, detection model and detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115690566B (en) | 2024-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10635979B2 (en) | Category learning neural networks | |
| CN109710800B (en) | Model generation method, video classification method, device, terminal and storage medium | |
| CN111611924B (en) | Mushroom identification method based on deep migration learning model | |
| CN114492574A (en) | Pseudo label loss unsupervised countermeasure domain adaptive picture classification method based on Gaussian uniform mixing model | |
| CN109033107A (en) | Image search method and device, computer equipment and storage medium | |
| CN109871885A (en) | A kind of plants identification method based on deep learning and Plant Taxonomy | |
| CN113076994A (en) | Open-set domain self-adaptive image classification method and system | |
| CN114092742B (en) | Multi-angle-based small sample image classification device and method | |
| CN113761259A (en) | Image processing method and device and computer equipment | |
| CN114358188A (en) | Feature extraction model processing method, feature extraction model processing device, sample retrieval method, sample retrieval device and computer equipment | |
| CN111832580B (en) | SAR target recognition method combining less sample learning and target attribute characteristics | |
| CN114677687A (en) | ViT and convolutional neural network fused writing brush font type rapid identification method | |
| CN114821204A (en) | Meta-learning-based embedded semi-supervised learning image classification method and system | |
| CN114742224A (en) | Pedestrian re-identification method and device, computer equipment and storage medium | |
| CN115690566B (en) | Deep sea animal new species identification method based on deep migration cluster learning | |
| CN114676769A (en) | Visual transform-based small sample insect image identification method | |
| CN115994239A (en) | Prototype comparison learning-based semi-supervised remote sensing image retrieval method and system | |
| CN113192108B (en) | Man-in-loop training method and related device for vision tracking model | |
| CN114581769A (en) | Method for identifying houses under construction based on unsupervised clustering | |
| CN114782997A (en) | Pedestrian re-identification method and system based on multi-loss attention adaptive network | |
| CN113077009A (en) | Tunnel surrounding rock lithology identification method based on migration learning model | |
| CN112926368A (en) | Method and device for identifying obstacle | |
| CN115115868B (en) | Multi-mode collaborative scene recognition method based on triples | |
| CN116384439B (en) | Target detection method based on self-distillation | |
| CN116129198B (en) | Multi-domain tire pattern image classification method, system, medium and equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |