CN111091162A - Method, system, terminal and medium for finely classifying species - Google Patents
Method, system, terminal and medium for finely classifying species Download PDFInfo
- Publication number
- CN111091162A CN111091162A CN202010193950.6A CN202010193950A CN111091162A CN 111091162 A CN111091162 A CN 111091162A CN 202010193950 A CN202010193950 A CN 202010193950A CN 111091162 A CN111091162 A CN 111091162A
- Authority
- CN
- China
- Prior art keywords
- image
- animal
- global
- classification
- classified
- 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
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a system, a terminal and a medium for finely classifying species, which comprises the following steps: acquiring an image of an animal to be classified; the fine classification convolutional neural network model carries out species fine classification identification on the animal image to be classified; and outputting the animal species classification recognition result by the fine classification convolutional neural network model. The invention has the beneficial effects that: the computer is used for automatically identifying the subclasses of the animals according to the characteristics of the animals in the images to be classified, so that the manual workload is reduced, and the accuracy is high.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to a method, a system, a terminal and a medium for finely classifying species based on a self-adaptive searching judgment area.
Background
The fine species classification task refers to more accurate subclassing of a large class of species, such as identifying different types of birds, flowers, vehicles, airplanes and the like. In real life, there is a wide application need to identify different subclasses. For example, in ecological environment protection, it is useful to effectively identify the subclass class of animals and plants to promote ecological research. However, this task presents certain challenges. On one hand, the subclasses in the same species have slight differences and are often distinguished by means of the local discriminant differences of the objects; on the other hand, the subclass has large intra-class difference, and has a plurality of influence factors such as attitude, shielding, background interference and the like. In order to solve the above problems, some researches have been conducted to distinguish different subclasses by using artificially labeled information, such as labeling the head, wings and tail of birds, which are considered to be distinguishable by human beings, and to identify the subclasses by combining the characteristics of the parts. However, such methods have the following disadvantages: 1. and a large amount of human resources are needed for region information labeling. 2. Regions that are considered by humans to be distinctive are not necessarily suitable for computer recognition.
Disclosure of Invention
Aiming at the defects in the prior art, the embodiment of the invention provides a species subdivision classification system, a terminal and a medium based on a self-adaptive searching and distinguishing area, which utilize a computer to automatically identify the specific class of an animal from the animal characteristics in an image to be classified, reduce the manual workload and have high accuracy.
In a first aspect, an embodiment of the present invention provides a method for fine classification of species based on adaptive search of a discriminant area, including the following steps:
acquiring an image of an animal to be classified;
the fine classification convolutional neural network model carries out species fine classification identification on the animal image to be classified;
and outputting the animal species classification recognition result by the fine classification convolutional neural network model.
Optionally, the specific method for identifying the fine species of the animal image to be classified by the fine-classification convolutional neural network model comprises the following steps:
sequentially carrying out full convolution network, global average pooling, full connection and softmax operation on the animal images to be classified to sequentially obtain a feature map based on the global images, global image features, scores of the global images belonging to the animal categories and classification results based on the global images;
generating a discriminant map according to the feature map and the classification weight information of the global image, wherein the discriminant map reflects the discriminant of a relevant region in the animal image to be classified;
selecting a distinguishing area from an animal image to be classified, and processing the distinguishing area to obtain area characteristics and a corresponding classification result;
and selecting partial regional features from the regional features to be spliced with the global image features to obtain mixed features, and performing softmax operation on the mixed features to obtain a prediction classification result.
Optionally, the specific method for obtaining the classification result based on the global image by sequentially performing full convolution network, global average pooling, full connection and softmax operations on the animal image to be classified includes:
the animal image to be classified is subjected to a full convolution network to obtain a feature map based on a global image;
carrying out global averaging on the feature map to obtain global image features;
obtaining the score of the global image belonging to the animal category through full connection of the global image features;
and performing softmax operation on the animal class scores to obtain a classification result based on the global image.
Optionally, the specific method for obtaining the region features and the corresponding classification results by selecting the discrimination regions from the images of the animals to be classified and processing the discrimination regions includes:
sorting the regions with discriminant according to the size of the pixel values of the discriminant image;
processing by adopting a non-maximum inhibition method, and selecting the first m discriminant areas, wherein m is an integer;
and cutting m regional images with discriminativity from the animal images to be classified, amplifying the regional images to the same size, and sequentially performing full convolution network, global average pooling, full connection and softmax operation on the cut images to obtain regional characteristics and corresponding classification results.
In a second aspect, the system for fine classification of species based on adaptive search for a discriminant region provided by the embodiments of the present invention includes an image acquisition module, a fine classification identification module, and an identification result output module,
the image acquisition module is used for acquiring an image of an animal to be classified;
the fine classification identification module is used for carrying out species fine classification identification on the animal image to be classified by utilizing a convolutional neural network model;
the identification result output module is used for outputting an animal species classification identification result.
Optionally, the detailed method for identifying the species sub-categories of the animal image to be classified by the sub-category identification module by using the convolutional neural network model comprises the following steps: sequentially carrying out full convolution network, global average pooling, full connection and softmax operation on the animal images to be classified to sequentially obtain a feature map based on the global images, global image features, scores of the global images belonging to the animal categories and classification results based on the global images;
generating a discriminant map according to the feature map and the classification weight information of the global image, wherein the discriminant map reflects the discriminant of a relevant region in the animal image to be classified;
selecting a region with high discriminability from the animal image to be classified according to the discriminant map, cutting the discriminant region from the original image, and sending the region to a convolutional neural network model to obtain region characteristics and a corresponding classification result;
and selecting partial regional features from the regional features to be spliced with the global image features to obtain mixed features, and performing softmax operation on the mixed features to obtain a prediction classification result.
Optionally, the fine classification recognition module sequentially performs full convolution network, global average pooling, full connection and softmax operations on the animal image to be classified, and the specific method for obtaining the classification result based on the global image includes:
the animal image to be classified is subjected to a full convolution network to obtain a feature map based on a global image;
carrying out global averaging on the feature map to obtain global image features;
obtaining the score of the global image belonging to the animal category through full connection of the global image features;
and performing softmax operation on the animal class scores to obtain a classification result based on the global image.
Optionally, the specific method for obtaining the region features and the corresponding classification results by the fine classification recognition module selecting the discrimination region and processing the discrimination region includes:
sorting the regions with discriminant according to the size of the pixel values of the discriminant image;
processing by adopting a non-maximum inhibition method, and selecting the first m discriminant areas, wherein m is an integer;
and cutting m regional images with discriminativity from the animal images to be classified, amplifying the regional images to the same size, and sequentially performing full convolution network, global average pooling, full connection and softmax operation on the cut images to obtain regional characteristics and corresponding classification results.
In a third aspect, an intelligent terminal provided in an embodiment of the present invention includes a processor, an input device, an output device, and a memory, where the processor, the input device, the output device, and the memory are connected to each other, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the method steps described in the foregoing embodiment.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, in which a computer program is stored, the computer program comprising program instructions, which, when executed by a processor, cause the processor to perform the method steps described in the above embodiments.
The invention has the beneficial effects that:
according to the method, the terminal and the medium for finely classifying the species based on the self-adaptive searching and judging area, provided by the embodiment of the invention, the subclasses of the animals are automatically identified by utilizing the characteristics of the animals in the image to be classified by utilizing the computer, so that the manual workload is reduced, and the accuracy is high.
Drawings
In order to more clearly illustrate the detailed description of the invention or the technical solutions in the prior art, the drawings that are needed in the detailed description of the invention or the prior art will be briefly described below. Throughout the drawings, like elements or portions are generally identified by like reference numerals. In the drawings, elements or portions are not necessarily drawn to scale.
Fig. 1 is a flowchart illustrating a method for fine classification of species based on adaptive search for a discriminant area according to a first embodiment of the present invention;
FIG. 2 illustrates a block diagram of a model of a fine-classification convolutional neural network of a first embodiment of the present invention;
fig. 3 is a schematic structural diagram illustrating a species subdivision classification system based on adaptive search for a discriminant area according to a second embodiment of the present invention;
fig. 4 shows a block diagram of an intelligent terminal according to a third 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.
As used in this specification and the appended claims, the term "if" may be interpreted contextually as "when", "upon" or "in response to a determination" or "in response to a detection". Similarly, the phrase "if it is determined" or "if a [ described condition or event ] is detected" may be interpreted contextually to mean "upon determining" or "in response to determining" or "upon detecting [ described condition or event ]" or "in response to detecting [ described condition or event ]".
It is to be noted that, unless otherwise specified, technical or scientific terms used herein shall have the ordinary meaning as understood by those skilled in the art to which the invention pertains.
Fig. 1 shows a flowchart of a fine species classification method based on adaptive search for a discriminant area according to a first embodiment of the present invention, where the method includes the following steps:
and S1, acquiring an image of the animal to be classified.
And S2, carrying out species fine classification identification on the animal image to be classified by the fine classification convolutional neural network model.
And S3, outputting the animal species classification recognition result by the fine classification convolutional neural network model.
The fine classification convolutional neural network model carries out fine classification recognition on animal species, and the main idea is to find out an area with discriminability in an image by using classification weight information of a global image. A frame diagram of a fine-classification convolutional neural network model is shown in fig. 2. Giving an input picture I, and obtaining a feature map based on a global image after passing through a full convolution network
Wherein h represents the width based on the global image feature map, w represents the length based on the global image feature map, and the feature map Z is subjected to global average pooling to obtain the global image feature
The score S of the global image belonging to each animal category is obtained through full connection of the features, and the calculation mode is as follows:
wherein C represents the total class of classification,
indicate full connectivity of
The weight of the class. And performing softmax operation on the category score S to obtain a classification result based on the global image.
Then, the feature map of the global image
And classification weight
For generating a discriminant map
The discrimination map M can reflect the correlation in the input image IThe discrimination of the region is generated as follows:
wherein max represents the number of pairs of C
And selecting the maximum value of the cross channel. Each pixel position in the determination map M corresponds to a certain position of the original image, and represents a certain area size. By performing the inverse estimation of the convolution operation perception area, the position and area size corresponding to the original image can be obtained, i.e. each pixel in the discriminant map M corresponds to an area position in the image I, and each pixel value can indicate the discriminability of the area (the higher the pixel value, the higher the discriminability). In order to select the first m regions with the most discriminant, the corresponding regions are sorted according to the pixel values, then the redundancy of the regions is reduced by adopting non-maximum suppression, and finally the first m regions with the most discriminant are selected. Cutting the areas from the original image I, amplifying the areas to the same size, and sequentially performing full convolution network, global average pooling, full connection and softmax operation to obtain the characteristics of the areas
And corresponding classification results.
To integrate global image features with regional features, a pre-selection is made from m regional features
Splicing the individual image with the global image characteristics to obtain mixed characteristics
The feature obtains its classification result through softmax operation.
In the network training phase, all softmax classification results are subjected to error gradient back-transmission by using a cross entropy loss function. All the full convolution networks share the weight, and the discriminant areas share the classification weight. In the testing stage, an image is input, and the output result of the mixed feature F is used as a final prediction classification result.
According to the method for finely classifying species based on the self-adaptive searching and distinguishing area, which is provided by the embodiment of the invention, the subclasses of the animals are automatically identified by utilizing the animal characteristics in the image to be classified by utilizing the computer, so that the manual workload is reduced, and the accuracy is high.
In the first embodiment described above, a species fine classification method based on adaptive finding of a discriminant area is provided, and correspondingly, the present application also provides a species fine classification system based on adaptive finding of a discriminant area. Please refer to fig. 3, which is a schematic structural diagram of a fine classification system based on adaptive search for a discriminant area according to a second embodiment of the present invention. Since the apparatus embodiments are substantially similar to the method embodiments, they are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for relevant points. The device embodiments described below are merely illustrative.
As shown in fig. 3, a fine species classification system based on adaptive search for a determination area according to a second embodiment of the present invention includes an image obtaining module, a fine classification identification module, and an identification result output module, where the image obtaining module is configured to obtain an image of an animal to be classified; the fine classification identification module is used for carrying out species fine classification identification on the animal image to be classified by utilizing the convolutional neural network model; and the identification result output module is used for outputting the animal species classification identification result.
Specifically, the method for identifying the fine species of the animal image to be classified by using the convolutional neural network model comprises the following steps: sequentially carrying out full convolution network, global average pooling, full connection and softmax operation on the animal images to be classified to sequentially obtain a feature map based on the global images, global image features, scores of the global images belonging to the animal categories and classification results based on the global images; generating a discriminant map according to the feature map and the classification weight information of the global image, wherein the discriminant map reflects the discriminant of a relevant region in the animal image to be classified; selecting a distinguishing area from an animal image to be classified, and processing the distinguishing area to obtain area characteristics and a corresponding classification result; and selecting partial regional features from the regional features to be spliced with the global image features to obtain mixed features, and performing softmax operation on the mixed features to obtain a prediction classification result.
Specifically, the method for obtaining the classification result based on the global image includes the following steps that a fine classification identification module in the system sequentially carries out full convolution network, global average pooling, full connection and softmax operation on an animal image to be classified: the animal image to be classified is subjected to a full convolution network to obtain a feature map based on a global image; carrying out global averaging on the feature map to obtain global image features; obtaining the score of the global image belonging to the animal category through full connection of the global image features; and performing softmax operation on the animal class scores to obtain a classification result based on the global image.
Specifically, the specific method for obtaining the region characteristics and the corresponding classification results by selecting the judgment region from the animal image to be classified by the fine classification identification module in the system and processing the judgment region comprises the following steps: sorting the regions with discriminant according to the size of the pixel values of the discriminant image; processing by adopting a non-maximum inhibition method, and selecting the first m discriminant areas, wherein m is an integer; and cutting m regional images with discriminativity from the animal images to be classified, amplifying the regional images to the same size, and sequentially performing full convolution network, global average pooling, full connection and softmax operation on the cut images to obtain regional characteristics and corresponding classification results.
The system for finely classifying species based on the self-adaptive searching for the judgment area and the method for finely classifying species based on the self-adaptive searching for the judgment area provided by the invention have the same inventive concept and the same beneficial effects, and are not repeated herein.
As shown in fig. 4, a schematic diagram of an intelligent terminal according to a third embodiment of the present invention is provided, where the terminal includes a processor, an input device, an output device, and a memory, where the processor, the input device, the output device, and the memory are connected to each other, the memory is used for storing a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the method described in the foregoing embodiment.
It should be understood that in the embodiments of the present invention, the Processor may be a Central Processing Unit (CPU), and the Processor 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. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The input device may include a touch pad, a fingerprint sensor (for collecting fingerprint information of a user and direction information of the fingerprint), a microphone, etc., and the output device may include a display (LCD, etc.), a speaker, etc.
The memory may include both read-only memory and random access memory, and provides instructions and data to the processor. The portion of memory may also include non-volatile random access memory. For example, the memory may also store device type information.
In a specific implementation, the processor, the input device, and the output device described in the embodiments of the present invention may execute the implementation described in the method embodiments provided in the embodiments of the present invention, and may also execute the implementation described in the system embodiments in the embodiments of the present invention, which is not described herein again.
The invention also provides an embodiment of a computer-readable storage medium, in which a computer program is stored, which computer program comprises program instructions that, when executed by a processor, cause the processor to carry out the method described in the above embodiment.
The computer readable storage medium may be an internal storage unit of the terminal described in the foregoing embodiment, for example, a hard disk or a memory of the terminal. The computer readable storage medium may also be an external storage device of the terminal, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the terminal. Further, the computer-readable storage medium may also include both an internal storage unit and an external storage device of the terminal. The computer-readable storage medium is used for storing the computer program and other programs and data required by the terminal. The computer readable storage medium may also be used to temporarily store data that has been output or is to be output.
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.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the terminal and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed terminal and method can be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may also be an electric, mechanical or other form of connection.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the present invention, and they should be construed as being included in the following claims and description.
Claims (10)
1. A species subdivision classification method based on self-adaptive search discrimination area is characterized by comprising the following steps:
acquiring an image of an animal to be classified;
the fine classification convolutional neural network model carries out species fine classification identification on the animal image to be classified;
and outputting the animal species classification recognition result by the fine classification convolutional neural network model.
2. The method for sub-category classification of species based on self-adaptive search for discriminant area according to claim 1, wherein the sub-category convolutional neural network model is used for sub-category identification of species of the animal image to be classified, and comprises the following steps:
sequentially carrying out full convolution network, global average pooling, full connection and softmax operation on the animal images to be classified to sequentially obtain a feature map based on the global images, global image features, scores of the global images belonging to the animal categories and classification results based on the global images;
generating a discriminant map according to the feature map and the classification weight information of the global image, wherein the discriminant map reflects the discriminant of a relevant region in the animal image to be classified;
selecting a region with high discriminability from the animal image to be classified according to the discriminant map, cutting the discriminant region from the original image, and sending the region to a convolutional neural network model to obtain region characteristics and a corresponding classification result;
and selecting partial regional features from the regional features to be spliced with the global image features to obtain mixed features, and performing softmax operation on the mixed features to obtain a prediction classification result.
3. The fine species classification method based on the adaptive search discrimination area as claimed in claim 2, wherein the specific method for obtaining the classification result based on the global image by sequentially subjecting the animal image to be classified to full convolution network, global average pooling, full connection and softmax operation comprises:
the animal image to be classified is subjected to a full convolution network to obtain a feature map based on a global image;
carrying out global averaging on the feature map to obtain global image features;
obtaining the score of the global image belonging to the animal category through full connection of the global image features;
and performing softmax operation on the animal class scores to obtain a classification result based on the global image.
4. The fine species classification method based on the self-adaptive judgment area searching as claimed in claim 3, wherein the specific method for selecting the judgment area from the animal image to be classified and processing the judgment area to obtain the area features and the corresponding classification results comprises:
sorting the regions with discriminant according to the size of the pixel values of the discriminant image;
processing by adopting a non-maximum inhibition method, and selecting the first m discriminant areas, wherein m is an integer;
and cutting m regional images with discriminativity from the animal images to be classified, amplifying the regional images to the same size, and sequentially performing full convolution network, global average pooling, full connection and softmax operation on the cut images to obtain regional characteristics and corresponding classification results.
5. A fine species classification system based on self-adaptive search and discrimination area is characterized by comprising an image acquisition module, a fine classification identification module and an identification result output module,
the image acquisition module is used for acquiring an image of an animal to be classified;
the fine classification identification module is used for carrying out species fine classification identification on the animal image to be classified by utilizing a convolutional neural network model;
the identification result output module is used for outputting an animal species classification identification result.
6. The system for sub-category species identification based on adaptive search for areas of discrimination of claim 5 wherein said sub-category identification module uses a convolutional neural network model to identify sub-category species of the images of the animal to be classified, comprising: sequentially carrying out full convolution network, global average pooling, full connection and softmax operation on the animal images to be classified to sequentially obtain a feature map based on the global images, global image features, scores of the global images belonging to the animal categories and classification results based on the global images;
generating a discriminant map according to the feature map and the classification weight information of the global image, wherein the discriminant map reflects the discriminant of a relevant region in the animal image to be classified;
selecting a distinguishing area from an animal image to be classified, and processing the distinguishing area to obtain area characteristics and a corresponding classification result;
and selecting partial regional features from the regional features to be spliced with the global image features to obtain mixed features, and performing softmax operation on the mixed features to obtain a prediction classification result.
7. The system for fine classification of species based on the adaptive search discrimination area according to claim 6, wherein the fine classification recognition module sequentially performs full convolution network, global average pooling, full connection and softmax operations on the image of the animal to be classified, and the specific method for obtaining the classification result based on the global image comprises:
the animal image to be classified is subjected to a full convolution network to obtain a feature map based on a global image;
carrying out global averaging on the feature map to obtain global image features;
obtaining the score of the global image belonging to the animal category through full connection of the global image features;
and performing softmax operation on the animal class scores to obtain a classification result based on the global image.
8. The system for fine classification of species based on self-adaptive finding of the discriminant area as claimed in claim 7, wherein the fine classification recognition module selects the discriminant area from the image of the animal to be classified, and processes the discriminant area to obtain the area features and the corresponding classification results, and the specific method comprises:
sorting the regions with discriminant according to the size of the pixel values of the discriminant image;
processing by adopting a non-maximum inhibition method, and selecting the first m discriminant areas, wherein m is an integer;
and cutting m regional images with discriminativity from the animal images to be classified, amplifying the regional images to the same size, and sequentially performing full convolution network, global average pooling, full connection and softmax operation on the cut images to obtain regional characteristics and corresponding classification results.
9. An intelligent terminal comprising a processor, an input device, an output device and a memory, the processor, the input device, the output device and the memory being interconnected, the memory being adapted to store a computer program, the computer program comprising program instructions, characterized in that the processor is configured to invoke the program instructions to perform the method steps according to any of claims 1 to 4.
10. A computer-readable storage medium, characterized in that the computer storage medium stores a computer program comprising program instructions which, when executed by a processor, cause the processor to carry out the method steps according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010193950.6A CN111091162A (en) | 2020-03-19 | 2020-03-19 | Method, system, terminal and medium for finely classifying species |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010193950.6A CN111091162A (en) | 2020-03-19 | 2020-03-19 | Method, system, terminal and medium for finely classifying species |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111091162A true CN111091162A (en) | 2020-05-01 |
Family
ID=70400623
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010193950.6A Pending CN111091162A (en) | 2020-03-19 | 2020-03-19 | Method, system, terminal and medium for finely classifying species |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111091162A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112529020A (en) * | 2020-12-24 | 2021-03-19 | 携程旅游信息技术（上海）有限公司 | Animal identification method, system, equipment and storage medium based on neural network |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106096531A (en) * | 2016-05-31 | 2016-11-09 | 安徽省云力信息技术有限公司 | A kind of traffic image polymorphic type vehicle checking method based on degree of depth study |
-
2020
- 2020-03-19 CN CN202010193950.6A patent/CN111091162A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106096531A (en) * | 2016-05-31 | 2016-11-09 | 安徽省云力信息技术有限公司 | A kind of traffic image polymorphic type vehicle checking method based on degree of depth study |
Non-Patent Citations (2)
| Title |
|---|
| ZHICONG FENG 等: "Learning to Focus and Discriminate for Fine-Grained Classification", 《INTERNATIONAL CONFERENCE ON IMAGE PROCESSING》 * |
| 翁雨辰 等: "深度区域网络方法的细粒度图像分类", 《中国图象图形学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112529020A (en) * | 2020-12-24 | 2021-03-19 | 携程旅游信息技术（上海）有限公司 | Animal identification method, system, equipment and storage medium based on neural network |
| CN112529020B (en) * | 2020-12-24 | 2024-05-24 | 携程旅游信息技术（上海）有限公司 | Animal identification method, system, equipment and storage medium based on neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108776819A (en) | A kind of target identification method, mobile terminal and computer readable storage medium | |
| CN109284733B (en) | Shopping guide negative behavior monitoring method based on yolo and multitask convolutional neural network | |
| CN109977191B (en) | Problem map detection method, device, electronic equipment and medium | |
| CN110443212B (en) | Positive sample acquisition method, device, equipment and storage medium for target detection | |
| CN108664840A (en) | Image-recognizing method and device | |
| CN110287936B (en) | Image detection method, device, equipment and storage medium | |
| CN111639653B (en) | False detection image determining method, device, equipment and medium | |
| CN113837257B (en) | Target detection method and device | |
| CN112561080A (en) | Sample screening method, sample screening device and terminal equipment | |
| CN107808126A (en) | Vehicle retrieval method and device | |
| CN111583180B (en) | Image tampering identification method and device, computer equipment and storage medium | |
| CN111709421A (en) | Bird identification method and device, computer equipment and storage medium | |
| CN109710788A (en) | Image pattern mark and management method and equipment | |
| CN113870254B (en) | Target object detection method and device, electronic equipment and storage medium | |
| CN113723467A (en) | Sample collection method, device and equipment for defect detection | |
| CN111523558A (en) | Ship shielding detection method and device based on electronic purse net and electronic equipment | |
| CN110210314B (en) | Face detection method, device, computer equipment and storage medium | |
| CN112541394A (en) | Black eye and rhinitis identification method, system and computer medium | |
| CN113454649B (en) | Target detection method, apparatus, electronic device, and computer-readable storage medium | |
| CN111091162A (en) | Method, system, terminal and medium for finely classifying species | |
| US7643674B2 (en) | Classification methods, classifier determination methods, classifiers, classifier determination devices, and articles of manufacture | |
| CN114140663A (en) | Multi-scale attention and learning network-based pest identification method and system | |
| CN109598712A (en) | Quality determining method, device, server and the storage medium of plastic foam cutlery box | |
| CN108985350B (en) | Method and device for recognizing blurred image based on gradient amplitude sparse characteristic information, computing equipment and storage medium | |
| CN111163332A (en) | Video pornography detection method, terminal and 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: 20200501 |
|
| RJ01 | Rejection of invention patent application after publication |