CN111814591A - Plant leaf identification system based on generative confrontation network model and IOS platform - Google Patents
Plant leaf identification system based on generative confrontation network model and IOS platform Download PDFInfo
- Publication number
- CN111814591A CN111814591A CN202010563204.1A CN202010563204A CN111814591A CN 111814591 A CN111814591 A CN 111814591A CN 202010563204 A CN202010563204 A CN 202010563204A CN 111814591 A CN111814591 A CN 111814591A
- Authority
- CN
- China
- Prior art keywords
- image
- plant leaf
- identification
- ios
- client
- 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
- 238000000034 method Methods 0.000 claims abstract description 33
- 230000003993 interaction Effects 0.000 claims abstract description 29
- 241000196324 Embryophyta Species 0.000 claims description 89
- 238000012545 processing Methods 0.000 claims description 28
- 238000012549 training Methods 0.000 claims description 21
- 230000008569 process Effects 0.000 claims description 12
- 238000013500 data storage Methods 0.000 claims description 10
- 238000004891 communication Methods 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 4
- 230000002452 interceptive effect Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 description 9
- 238000013527 convolutional neural network Methods 0.000 description 7
- 238000013135 deep learning Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 230000018109 developmental process Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 230000002902 bimodal effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007797 corrosion Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000010076 replication Effects 0.000 description 2
- 241000894007 species Species 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 208000032170 Congenital Abnormalities Diseases 0.000 description 1
- 238000013256 Gubra-Amylin NASH model Methods 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000003042 antagnostic effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 235000013311 vegetables Nutrition 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
- H04L41/0668—Management of faults, events, alarms or notifications using network fault recovery by dynamic selection of recovery network elements, e.g. replacement by the most appropriate element after failure
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Computer Hardware Design (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a plant leaf identification system based on a generative confrontation network model and an IOS platform, which comprises an IOS client and a server side which is connected with the IOS client through a wireless network, wherein the IOS client acquires a plant leaf image and preprocesses the plant leaf image, the processed plant leaf image selects a local identification path or a server side identification path through a man-machine interaction page of the IOS client, the IOS client requests the local identification path to call a self network model for plant leaf identification, and the server side calls the generative confrontation network based on segment loss weighting to the preprocessed plant leaf image sent by the IOS client in a wireless mode to identify after receiving a request of the server side identification path. Compared with the prior art, the method has the advantages of improving the mode collapse phenomenon of the model, improving the recognition efficiency and the like.
Description
Technical Field
The invention relates to the technical field of plant leaf identification, in particular to a plant leaf identification system based on a generative confrontation network model and an IOS platform.
Background
There are many plant species on earth, and since there are many kinds of plants, it is difficult to memorize and recognize many kinds of plants unless there is professional knowledge about them, so that the research of the plant automatic recognition technology is of great significance. The leaves are important organs of plants and can reflect important characteristics of the plants, so the leaves are often used as important judgment bases for plant identification.
At present, there are three main methods for identifying plant species:
(1) artificially attached species label: the method is a method adopted by most of vegetable gardens or parks for facilitating tourists to identify plant species, namely, label plates engraved with plant related information are attached to plant branches for the tourists to read. The method has the congenital defects of consuming manpower and material resources, less information transmission, unobtrusive expression, easy corrosion of the label and the like, and is only popularized in charged or protected scenic spots;
(2) two-dimensional code on artifical subsides: the method can be regarded as an upgrade version of the method (1), is a product combining the modern electronic technology development of species labels, tourists can obtain rich plant species information by scanning two-dimensional codes attached to plant branches and accessing the Internet, the method overcomes the defect that the transmitted information in the method (1) is less, but still has the defects of manpower and material resource consumption, easy corrosion of labels and the like, is implemented only in a few gardens and is still in a test stage at present;
(3) study by professional plant taxonomy workers: this is the most traditional research method of plant taxonomy, and researchers are through collecting the sample and artifical the measurement, and combine experience knowledge and books to guide to classify the sample, and this kind of method work load is huge, and needs a large amount of professional knowledge, can only implement in scientific research field.
The three methods cannot be popularized due to the defects of the three methods, and no convenient, quick and low-cost plant species identification method exists in the market at present.
The image recognition problem is an important problem in the field of computer vision, and the traditional image recognition method mainly comprises the steps of extracting various features based on an original image, and inputting the obtained features into various classifiers for recognition. The problems of different visual angles, size change of an object, deformation of the object, shielding, illumination intensity change and the like are involved in image identification, and manual design of image feature description is difficult to use, so that the identification accuracy of the traditional method is generally low. The convolutional neural network is a feedforward neural network, comprises a plurality of convolutional layers, pooling layers, full-link layers and other network layer structures, can fully utilize a two-dimensional structure of input data, has excellent performance in the image field, and is more suitable for solving related problems such as image recognition compared with other deep learning structures.
Deep learning has enjoyed great success in many fields and has received much attention. The reason why the neural network can be revived again is several. First, the appearance of large data has greatly alleviated the problem of overfitting in training. For example, the ImageNet training set has millions of annotated images. Secondly, the rapid development of various hardware provides very strong computing power, and even hardware devices specially used for deep learning appear, so that the training of a large-scale neural network becomes possible. Finally, in recent years, many research achievements appear in aspects of network model architecture, neural network parameter initialization, model training method, activation function selection and the like, so that the parameters of the network model can be reduced while the depth network model tends to be in depth, and the training speed and the recognition accuracy of the depth network model are greatly improved.
In recent years, with the rapid development of deep learning technology, relevant workers at home and abroad have achieved abundant results in the field of plant leaf identification. For example, the plant leaf is identified by using two characteristics of texture and shape, the texture characteristic is extracted by using Gabor filtering and gray level co-occurrence matrix, and the shape characteristic of the leaf edge is extracted by using Curvelet transform coefficient and invariant matrix. As another example, a Probabilistic Neural Network (PNN) based image processing method inputs PNN by extracting image features and orthogonally transforming into five variables. For another example, a plurality of different convolutional neural network models are adopted to perform recognition training on the plant, or a CNN + SVM model and a CNN + Softmax model are used, or a CNN + SVM model is used, and a 16-layer VGG-Net convolutional neural network model is adopted to perform training and testing on the plant sample.
Although deep learning is well performed on labeled data in the field of plant leaf identification, data marking costs a lot of manpower and material resources, how to adopt semi-supervised learning, and save manpower cost without influencing experimental results, and the use of unlabeled data is very worthy of research. In the field of plant leaf identification, sample collection of part of plants is very difficult, but a traditional deep learning method such as a convolutional neural network can only judge the class of the plant, but cannot supplement or train a data set, and how to supplement the training data set by using generated data is very important for reducing development cost. Therefore, it is important to develop a recognition system applied to plant leaves based on deep learning.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a plant leaf identification system based on a generative confrontation network model and an IOS platform.
The purpose of the invention can be realized by the following technical scheme:
the plant leaf identification system based on the generative confrontation network model and the IOS platform comprises an IOS client and a server, wherein the server and the IOS client are connected with each other through a wireless network. The IOS client side acquires a plant leaf image, preprocesses the plant leaf image, selects a local identification path or a server side identification path from the processed plant leaf image through a human-computer interaction page of the IOS client side, and calls a self network model to the local identification path to identify the plant leaf; and after receiving the request of identifying the path by the server, the server calls a generative countermeasure network based on the segment loss weighting to the preprocessed plant leaf image wirelessly sent by the IOS client for identification.
The IOS client comprises:
the system comprises an image acquisition module, an image uploading module, a client data storage module, a human-computer interaction page and a client network communication module, wherein a human-computer interaction interface is respectively connected with the image acquisition module and the image uploading module, an image processing module is respectively connected with the image acquisition module, the image uploading module, an image recognition module, the human-computer interaction interface, the client data storage module and the client network communication module, the image uploading module is connected with the client data storage module, and the client network communicator is connected with a server. The image processing module is used for processing the plant leaf images of the image acquisition module and the image uploading module, and the image identification module is used for rapidly identifying the local images of the IOS client.
The server side comprises a traffic distribution server of the extranet and a plurality of working servers used for respectively distributing the requests according to a certain distribution rule, each working server is respectively provided with a corresponding subordinate backup server, and the traffic distribution server is provided with a backup server.
The requests sent by the users are firstly all forwarded to a high-performance traffic distribution server, and the high-performance traffic distribution server distributes the user requests to relatively idle working servers for processing according to the running state of each server in the current server cluster so as to maintain the whole cluster in a relatively balanced state. On the other hand, in order to maintain the fault tolerance of the cluster, namely, the whole cluster can still normally operate under the condition that part of the working servers have faults, the fault tolerance support is provided for the high-performance working servers in the cluster through a master-slave replication technology, namely, a slave backup server is equipped for the high-performance traffic distribution server responsible for requesting forwarding in the cluster, the slave backup server is responsible for monitoring the running state of the high-performance traffic distribution server responsible for requesting forwarding, and when the high-performance traffic distribution server responsible for requesting forwarding has a machine fault, the slave backup server starts to take over the relevant work of requesting forwarding, so that the fault tolerance of the whole cluster is improved.
The specific steps of the system for identifying the plant leaves comprise:
1) the IOS client side obtains the blade image and carries out the complex background removing process of the image in an interactive mode. Specifically, the IOS client selects a locally stored plant leaf image for plant leaf image acquisition by shooting the plant leaf or the image, and the acquired image is uploaded to the image processing module for segmentation processing of the complex background.
Further, the IOS client side obtains the leaf image and carries out plant leaf image background segmentation processing on the image by adopting an SRN-DeblurNet network structure.
2) And (3) selecting an identification path on a man-machine interaction page by a user according to requirements, executing the step 3) if selecting a local quick identification step, and executing the step 4) if selecting a server-side identification step.
3) After the user selects local quick identification, the man-machine interaction page issues a control signal to the image identification module, the image identification module calls the deployed model to directly identify the plant leaf image processed in the step 1), and the identification result is displayed through the man-machine interaction page.
Further, the image recognition module carries out recognition through a lightweight network model mobileNet deployed on the IOS client.
4) After the user selects the server to identify, the man-machine interaction page sends the control signal to the image processing module, the image processing module compresses the picture, then the man-machine interaction page sends a request to the server, and the compressed picture is transmitted to the server through a wireless network. The specific contents of compressing the picture are as follows:
selecting a threshold, compressing the plant leaf image by adopting SR, and when the Pre PSNR is larger than a preset threshold, downsampling and decoding the image and then carrying out SRCNN filtering.
5) And after receiving the pictures transmitted by the IOS client, the server calls a generative confrontation network model based on the segment loss weighting and deployed on the server to identify the plant leaves, and returns the result to the man-machine interaction page of the IOS client. The specific contents of the plant leaf identification based on the generative confrontation network model with the segment loss weighting are as follows:
and controlling the GAN to adopt different forms of losses in different training stages, wherein the loss function in the second form is taken as the main part in the early training stage, the real sample and the generated sample are overlapped in the training process, and when the overlap reaches a switching parameter point, the switching is carried out mainly by taking the loss function in the first form.
Compared with the prior art, the invention has the following beneficial effects:
1) the system of the invention utilizes the picture classification capability of a semi-supervised generation type countermeasure model, and leads the generator to adopt different loss functions in different training stages by introducing time parameters, so that JS divergence can play a positive role; in order to provide enough gradient for a generator, extra characteristic-level mean square error loss and countermeasure loss are introduced for weighting, the model is used for semi-supervised image classification, mode collapse can be avoided to a certain extent, and a good recognition effect is achieved, so that plant leaf pictures uploaded by a client are effectively recognized, and a recognition result is accurately returned;
2) the method is provided with a local quick identification function, and by using a lightweight model of mobileNet and utilizing the image classification effect of a mobileNet classifier, the local plant leaf image can be quickly identified, so that the identification efficiency is high;
3) for the accurate plant leaf identification process at the server end, a generation type confrontation network based on the segmented loss weighting is applied at the server end, the training process is more stable by changing the training process of a generator and a discriminator and introducing the characteristic level loss between a real sample and a generated sample, and the mode collapse phenomenon of a model can be improved to a certain extent; and the model improves the performance of the discriminator and makes the extracted features more robust.
Drawings
FIG. 1 is a schematic flow chart of the plant leaf identification realized by the system of the present invention;
FIG. 2 is a schematic diagram of a server according to the present invention;
fig. 3 is a flow chart illustrating identification of a generative countermeasure network based on segment loss weighting.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.
The invention relates to a plant leaf identification system based on a generative confrontation network model and an IOS platform, which is used for identifying plant species and feeding back identification according to images of plant leaves.
The IOS client comprises an image acquisition module, an image uploading module, an image processing module, an image recognition module, a human-computer interaction interface, a client data storage module and a client network communication module, wherein the human-computer interaction interface is connected with the image acquisition module and the image uploading module, the image processing module is respectively connected with the image acquisition module, the image uploading module, the human-computer interaction interface, the client data storage module and the client network communication module, the image processing module is connected with the image recognition module, the image uploading module is connected with the client data storage module, and the client network communicator is connected with the server. The image processing module is used for preprocessing the local pictures (namely the plant leaf images of the image acquisition module and the image uploading module) of the IOS client. The image recognition module is used for rapidly recognizing the local images of the IOS client.
And the server side is used for accurately identifying the plant leaf images sent by the IOS client side. The server side comprises a flow distribution server, a backup server and a plurality of working servers, the backup server is connected with the flow distribution server, and the working servers are respectively connected with the flow distribution server.
The requests sent by the users are firstly all forwarded to a high-performance traffic distribution server, and the high-performance traffic distribution server distributes the user requests to relatively idle working servers for processing according to the running state of each server in the current server cluster so as to maintain the whole cluster in a relatively balanced state. On the other hand, in order to maintain the fault tolerance of the cluster, namely, the whole cluster can still normally operate under the condition that part of the working servers have faults, the fault tolerance support is provided for the high-performance working servers in the cluster through a master-slave replication technology, namely, a slave backup server is equipped for the high-performance traffic distribution server responsible for requesting forwarding in the cluster, the slave backup server is responsible for monitoring the running state of the high-performance traffic distribution server responsible for requesting forwarding, and when the high-performance traffic distribution server responsible for requesting forwarding has a machine fault, the slave backup server starts to take over the relevant work of requesting forwarding, so that the fault tolerance of the whole cluster is improved.
As shown in FIG. 1, the specific steps of the system of the present invention for identifying plant leaves include:
step 1, the IOS client side obtains blade images and carries out a picture complex background removing process on the images in an interactive mode.
The IOS client can shoot plant leaves through the image acquisition module, can also select certain plant leaf images in the client data storage module through the image uploading module, and uploads the images to the image processing module for division processing of complex backgrounds after the images are shot or selected. Preferably, the present invention uses an SRN-DeblurNet network structure for the plant leaf image background segmentation process, which takes as input a sequence of blurred images down-sampled at different scales from the input image, and then obtains a set of corresponding sharp images. The sharp image under the full resolution is the final output, which is convenient for the subsequent processing and the picture clearness. The background segmentation process is a prior art and will not be described herein in detail.
And 2, the user can randomly select an identification way on the man-machine interaction page according to the requirement of the user. Namely, a local quick identification step or a server-side accurate identification step is selected.
And 3, after the user selects local quick identification, the man-machine interaction page sends a control signal to the image identification module, the image identification module directly identifies the plant leaves processed in the step 1 by calling the deployed model, and the identification result is quickly displayed on the man-machine interaction page.
And local quick identification, namely identification is carried out through a lightweight network model mobileNet deployed on the IOS client. The identification by using the lightweight network model mobileNet is the prior art, and is not described in detail herein.
And 4, accurately identifying the server, namely identifying the antagonistic network based on the generation formula of the segment loss weighting through a network model deployed on the server. When the user selects the server side for accurate identification, the man-machine interaction page issues the control signal to the image processing module, and the image processing module compresses the picture to ensure the high efficiency of data transmission. And then, sending a request by the man-machine interaction page, and transmitting the compressed image to a server side through a client side network communication module.
Specifically, the operation of picture compression is: if an obvious target and background exist in the image, the gray level histogram of the image is in bimodal distribution, and when the gray level histogram has bimodal characteristics, the gray level corresponding to the valley between two peaks is selected as a threshold value. If the gray value of the background can reasonably be seen as constant throughout the image and all objects have almost the same contrast to the background, then a threshold can be chosen and SR (super resolution) can be used for compressing the plant leaf image. When Pre PSNR (Peak Signal to Noise Ratio) is greater than a predetermined threshold, the image is downsampled to (0.5W, 0.5H) and SRCNN filtering is performed after decoding.
And 5, after receiving the pictures transmitted by the IOS client, the server calls the model deployed on the working server, identifies the model and returns the result. In terms of algorithm, the method adopts a generation type countermeasure network based on the segment loss weighting to carry out accurate identification. The network controls the GAN to adopt different forms of losses in different training stages on the basis of the traditional semi-supervised generation type countermeasure network. The second form loss function is taken as the main part in the early training period, the real sample and the generated sample can be overlapped with the training, and after the switching parameter point is reached, the switching is carried out to the first form loss function, and the JS divergence can play a good role at the moment, so that the gradient disappearance and the mode collapse of the generator are avoided. The flow chart of the specific recognition is shown in fig. 3. After background segmentation and morphological processing, feature extraction is carried out on the picture through a depth convolution layer, classification and identification are carried out through a classifier of a semi-supervised generation type countermeasure network, and then a classification result is output.
After the user selects local quick identification, the system calls a model deployed at the mobile phone end to directly identify the plant leaves and quickly display the identification result. When the user selects the cloud accurate identification function, the system compresses the picture to ensure the high efficiency of data transmission. And then sending a request to transmit the picture to a server side. After receiving the pictures transmitted by the client, the server calls the model deployed on the server, identifies the models and returns results through a network protocol. The mechanism for supervising learning and loss weighting has obvious effect when acting on a deep GAN model. The model has a good identification effect on an ICL plant leaf data set, and the accuracy is improved by 4.77% compared with a ResNet50 basic network model.
The final experiment result shows that the method can effectively process the pictures submitted by the IOS client, return the identified accurate identification result to the human-computer interaction page of the IOS client, and improve the identification precision of the plant leaves.
The system of the invention utilizes the picture classification capability of a semi-supervised generation type countermeasure model, and leads the generator to adopt different loss functions in different training stages by introducing time parameters, so that JS divergence can play a positive role; in order to give enough gradient to the generator, extra feature level mean square error loss and countermeasure loss are introduced for weighting, the model is used for semi-supervised image classification, mode collapse can be avoided to a certain extent, a good recognition effect is achieved, plant leaf pictures uploaded by a client are effectively recognized, and a recognition result is accurately returned.
While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and those skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. The plant leaf identification system based on the generative confrontation network model and the IOS platform is characterized by comprising an IOS client and a server side which is connected with the IOS client through a wireless network, wherein the IOS client acquires a plant leaf image and preprocesses the plant leaf image, the processed plant leaf image selects a local identification path or a server side identification path through a man-machine interaction page of the IOS client, the IOS client requests the local identification path to call the network model for plant leaf identification, and the server side calls the generative confrontation network based on segment loss weighting to the preprocessed plant leaf image sent by the IOS client in a wireless mode to identify after receiving the request of the server side identification path.
2. The plant leaf identification system based on the generative countermeasure network model and the IOS platform as claimed in claim 1, wherein the IOS client is provided with an image processing module for processing the plant leaf image of the image acquisition module and the image uploading module and an image identification module for rapidly identifying the local picture of the IOS client.
3. The plant leaf identification system based on the generative countermeasure network model and the IOS platform as claimed in claim 2, wherein the IOS client further comprises an image acquisition module, an image upload module, a client data storage module and a client network communication module, the human-computer interaction interface is connected with the image acquisition module and the image upload module respectively, the image processing module is connected with the image acquisition module, the image upload module, the image identification module, the human-computer interaction interface, the client data storage module and the client network communication module respectively, the image upload module is connected with the client data storage module, and the client network communicator is connected with the server.
4. The plant leaf identification system based on generative confrontation network model and IOS platform as claimed in claim 2, wherein the specific steps of the system for plant leaf identification comprises:
1) the IOS client side obtains the blade image and carries out a picture complex background removing process on the image in an interactive mode;
2) a user selects an identification path on a man-machine interaction page according to requirements, if a local quick identification step is selected, the step 3) is executed, and if a server-side identification step is selected, the step 4) is executed;
3) after the user selects local quick identification, the man-machine interaction page sends a control signal to the image identification module, the image identification module calls a deployed model to directly identify the plant leaf image processed in the step 1), and an identification result is displayed through the man-machine interaction page;
4) after the user selects the server to identify, the man-machine interaction page sends a control signal to the image processing module, the image processing module compresses the picture, then the man-machine interaction page sends a request to the server, and the compressed picture is transmitted to the server through a wireless network;
5) and after receiving the pictures transmitted by the IOS client, the server calls a generative confrontation network model based on the segment loss weighting and deployed on the server to identify the plant leaves, and returns the result to the man-machine interaction page of the IOS client.
5. The generative confrontation network model and IOS platform based plant leaf identification system as claimed in claim 4 wherein in step 3), the image recognition module performs recognition through a lightweight network model mobileNet deployed at IOS client.
6. The plant leaf identification system based on generative confrontation network model and IOS platform as claimed in claim 4, wherein in step 5), the specific content of the plant leaf identification based on generative confrontation network model weighted by segment loss is:
and controlling the GAN to adopt different forms of losses in different training stages, wherein the loss function in the second form is taken as the main part in the early training stage, the real sample and the generated sample are overlapped in the training process, and when the overlap reaches a switching parameter point, the switching is carried out mainly by taking the loss function in the first form.
7. The plant leaf identification system based on the generative confrontation network model and IOS platform as claimed in claim 4, wherein in step 4), the image processing module compresses the image by:
selecting a threshold, compressing the plant leaf image by adopting SR, and when the Pre PSNR is larger than a preset threshold, downsampling and decoding the image and then carrying out SRCNN filtering.
8. The plant leaf identification system based on the generative countermeasure network model and the IOS platform as claimed in claim 4, wherein in step 1), the IOS client obtains the leaf image and performs the plant leaf image background segmentation process on the image by using SRN-DeblurNet network structure.
9. The plant leaf identification system based on the generative countermeasure network model and the IOS platform as claimed in claim 4, wherein in step 1), the IOS client selects the locally stored plant leaf image by shooting the plant leaf or image to collect the plant leaf image, and the collected image is uploaded to the image processing module to perform the segmentation processing of the complex background.
10. The plant leaf identification system based on the generative countermeasure network model and the IOS platform as claimed in claim 1, wherein the server comprises a traffic distribution server for extranet and a plurality of working servers for respectively issuing requests according to a certain distribution rule, each working server is provided with a respective slave backup server, and the traffic distribution server is provided with a backup server.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2020102422893 | 2020-03-31 | ||
CN202010242289 | 2020-03-31 |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111814591A true CN111814591A (en) | 2020-10-23 |
Family
ID=72846537
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010563204.1A Pending CN111814591A (en) | 2020-03-31 | 2020-06-19 | Plant leaf identification system based on generative confrontation network model and IOS platform |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111814591A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112418082A (en) * | 2020-11-22 | 2021-02-26 | 同济大学 | Plant leaf identification system and method based on metric learning and depth feature learning |
CN112686205A (en) * | 2021-01-14 | 2021-04-20 | 电子科技大学中山学院 | Parameter updating method and device and multi-terminal network architecture |
CN112699805A (en) * | 2020-12-31 | 2021-04-23 | 上海市崇明区蔬菜科学技术推广站 | Intelligent recognition system for vegetable pest control |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102332992A (en) * | 2008-06-17 | 2012-01-25 | 华为技术有限公司 | Parlay application server, SIP (session initiation protocol) dispatcher upgrading method and device |
CN206236111U (en) * | 2016-10-14 | 2017-06-09 | 同济大学 | A kind of leaf image plant automatic identification equipment based on interactive voice |
CN108665058A (en) * | 2018-04-11 | 2018-10-16 | 徐州工程学院 | A kind of generation confrontation network method based on segmentation loss |
CN110070071A (en) * | 2019-04-30 | 2019-07-30 | 重庆文理学院 | A kind of ginger cauline leaf pest and disease damage recognition methods and system based on image recognition |
-
2020
- 2020-06-19 CN CN202010563204.1A patent/CN111814591A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102332992A (en) * | 2008-06-17 | 2012-01-25 | 华为技术有限公司 | Parlay application server, SIP (session initiation protocol) dispatcher upgrading method and device |
CN206236111U (en) * | 2016-10-14 | 2017-06-09 | 同济大学 | A kind of leaf image plant automatic identification equipment based on interactive voice |
CN108665058A (en) * | 2018-04-11 | 2018-10-16 | 徐州工程学院 | A kind of generation confrontation network method based on segmentation loss |
CN110070071A (en) * | 2019-04-30 | 2019-07-30 | 重庆文理学院 | A kind of ginger cauline leaf pest and disease damage recognition methods and system based on image recognition |
Non-Patent Citations (3)
Title |
---|
XIN TAO ET AL.: "Scale-recurrent Network for Deep Image Deblurring", 《CVPR2018》 * |
ZHENGXUE CHENG ET AL.: "Performance Comparison of Convolutional AutoEncoders, Generative Adversarial Networks and Super-Resolution for Image Compression", 《ARXIV》 * |
李林等: "《数字社区系统工程》", 31 March 2010 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112418082A (en) * | 2020-11-22 | 2021-02-26 | 同济大学 | Plant leaf identification system and method based on metric learning and depth feature learning |
CN112699805A (en) * | 2020-12-31 | 2021-04-23 | 上海市崇明区蔬菜科学技术推广站 | Intelligent recognition system for vegetable pest control |
CN112686205A (en) * | 2021-01-14 | 2021-04-20 | 电子科技大学中山学院 | Parameter updating method and device and multi-terminal network architecture |
CN112686205B (en) * | 2021-01-14 | 2023-10-13 | 电子科技大学中山学院 | Parameter updating method and device and multi-terminal network architecture |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107256246B (en) | printed fabric image retrieval method based on convolutional neural network | |
Xiong et al. | Learning to generate time-lapse videos using multi-stage dynamic generative adversarial networks | |
Xu et al. | Light-YOLOv3: fast method for detecting green mangoes in complex scenes using picking robots | |
CN111814591A (en) | Plant leaf identification system based on generative confrontation network model and IOS platform | |
CN112232425B (en) | Image processing method, device, storage medium and electronic equipment | |
CN104463117B (en) | A kind of recognition of face sample collection method and system based on video mode | |
CN110223292B (en) | Image evaluation method, device and computer readable storage medium | |
CN103927387B (en) | Image indexing system and its correlation technique and device | |
CN109543697A (en) | A kind of RGBD images steganalysis method based on deep learning | |
CN105095902B (en) | Picture feature extracting method and device | |
CN104063686B (en) | Crop leaf diseases image interactive diagnostic system and method | |
CN109033954A (en) | A kind of aerial hand-written discrimination system and method based on machine vision | |
CN111242019B (en) | Video content detection method and device, electronic equipment and storage medium | |
CN112163508A (en) | Character recognition method and system based on real scene and OCR terminal | |
CN104346630A (en) | Cloud flower identifying method based on heterogeneous feature fusion | |
CN112418082A (en) | Plant leaf identification system and method based on metric learning and depth feature learning | |
CN109598250A (en) | Feature extracting method, device, electronic equipment and computer-readable medium | |
CN112560865A (en) | Semantic segmentation method for point cloud under outdoor large scene | |
CN114663766A (en) | Plant leaf identification system and method based on multi-image cooperative attention mechanism | |
CN115410087A (en) | Transmission line foreign matter detection method based on improved YOLOv4 | |
CN111341462A (en) | Mobile phone terminal diabetic retinopathy screening APP based on deep learning | |
CN116485646A (en) | Micro-attention-based light-weight image super-resolution reconstruction method and device | |
CN109711324A (en) | Human posture recognition method based on Fourier transformation and convolutional neural networks | |
CN104331717A (en) | Feature dictionary structure and visual feature coding integrating image classifying method | |
CN110796650A (en) | Image quality evaluation method and device, 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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201023 |