CN111047551A - Remote sensing image change detection method and system based on U-net improved algorithm - Google Patents
Remote sensing image change detection method and system based on U-net improved algorithm Download PDFInfo
- Publication number
- CN111047551A CN111047551A CN201911073929.6A CN201911073929A CN111047551A CN 111047551 A CN111047551 A CN 111047551A CN 201911073929 A CN201911073929 A CN 201911073929A CN 111047551 A CN111047551 A CN 111047551A
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- image
- network
- net
- change detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000008859 change Effects 0.000 title claims abstract description 69
- 238000001514 detection method Methods 0.000 title claims abstract description 63
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 claims abstract description 41
- 238000004458 analytical method Methods 0.000 claims abstract description 24
- 238000005070 sampling Methods 0.000 claims abstract description 24
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 101100295091 Arabidopsis thaliana NUDT14 gene Proteins 0.000 claims abstract 7
- 238000012545 processing Methods 0.000 claims description 38
- 238000000605 extraction Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 8
- 238000011176 pooling Methods 0.000 claims description 7
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 claims description 6
- 238000013461 design Methods 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 6
- 230000006872 improvement Effects 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 abstract description 4
- 230000007246 mechanism Effects 0.000 abstract description 3
- 230000008447 perception Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 10
- 238000013135 deep learning Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 4
- 238000004880 explosion Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 230000009467 reduction Effects 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000008034 disappearance Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013529 biological neural network Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- 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
- G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- 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]
-
- 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
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a remote sensing image change detection method and a system based on a U-net improved algorithm, wherein the method comprises the following steps: preprocessing the remote sensing images, and then integrating the preprocessed remote sensing images in different time phases to obtain an input image; performing down-sampling coding and up-sampling decoding on an input image based on the improved U-net network; and finally, outputting a binary image of whether the image changes according to the analysis result. The invention applies the network structure of the semantic segmentation direction to the field of change detection, introduces a residual learning mechanism to enable an encoder to quickly converge and deepen the network layer number, and simultaneously uses the ASPP to enhance the perception capability of the network to the image characteristics, so that the algorithm can be kept at a higher level in both precision and efficiency and has stronger robustness. The method is suitable for the change detection field of remote sensing images, can be popularized to other fields, and has important significance.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to a remote sensing image change detection method and system based on a U-net improved algorithm.
Background
The remote sensing image change detection work means that the remote sensing images of different time phases of the same area are quantitatively analyzed aiming at the same area, so that the form and the characteristics of the earth surface in the image are determined, and the changed part and the unchanged part of the area at different time are detected. The remote sensing image change detection has a wide effect in the direction of land resource control, and is widely applied to the detection of the directions of building violation construction, natural disaster assessment, urban construction planning and the like.
With the continuous development of satellite technology, remote sensing images at the present stage have the characteristics of short cruising period, large data volume and the like, and the method for detecting regional changes by using manual contrast analysis cannot meet the requirements. How to quickly and accurately extract the change information of the corresponding area from the remote sensing image is one of the hot problems in the field of remote sensing image application at the present stage.
On the other hand, with the rapid development of device computing capability and the general improvement of data quality in recent years, deep learning becomes a popular direction in the field of computer device learning, and meanwhile, the deep learning has achieved excellent performances in the directions of computer vision, natural language processing and the like. The principle of deep learning is to simulate a biological neural network and build a network structure consisting of interconnected neurons. When data such as images and texts are input to the network, the network can extract deep features in the data for analysis through coordination among network layers, and finally the interpretation of the data is fed back.
The intermediate deep learning method is excellent in computer vision, the remote sensing image change detection also starts to be a method for constructing a differential image and then performing threshold division, the method is combined with a deep learning algorithm, a deep neural network is used for learning nonlinear change characteristics in satellite images in different time phases, and a final change detection result is determined. At present, a deep learning idea provides a new idea for satellite image processing, and a change detection algorithm based on deep learning becomes a new hotspot in the field of satellite image change detection.
① the change detection method facing to the pixel is mainly two types, ① the change detection method facing to the pixel, the thought is to take the pixel as the basic unit, analyze the pixel according to the spectrum and domain characteristics of the pixel, change the detection problem of the remote sensing image into the two classification problems of the target change and the non-change, the method takes each pixel as the unit to operate, thus it is easy to get many noise information, influence the precision, ② the change detection facing to the object, make an improvement on the method taking the pixel as the unit, first cluster a plurality of pixels, make it form an object, then classify the object, then compare the classification situation of the different time phase images to get the detection result.
One of the solutions is based on the change detection of the stack noise reduction auto-encoder network. The scheme belongs to a pixel-oriented change detection technology, and comprises the steps of firstly extracting the domain characteristics of a remote sensing image, then inputting the domain characteristics of each pixel point into a stack noise reduction automatic encoder, then training layer by layer, and carrying out supervised fine tuning until the training is completed. In the existing change detection technology, a proper neighborhood range is selected, and the method can achieve a good change detection effect on the basis of reducing noise. However, the method depends on the selection of the neighborhood range in practical application, and false detection and missed detection are easy to occur if the method is not selected properly. Meanwhile, the network structure is relatively simple, the number of layers is small, and therefore although the effect is better than that of the traditional method, new breakthrough is difficult to take place.
In summary, the mainstream methods for detecting changes in remote sensing images in the prior art are mainly classified into two types, i.e., first classification and then comparison, and first comparison and then analysis, where the former method corresponds to object-oriented change detection, and the latter method corresponds to pixel-oriented change detection. The existing remote sensing image change detection methods have some defects.
The object-oriented change detection method can use information such as texture features, shape features and neighborhood relations among objects as a basis for change detection, but classification is performed after pixel points are clustered, so that the precision of the ground classification greatly influences the precision of the change detection, and a serious error accumulation problem also exists. Meanwhile, the method has higher requirement on the quality of the remote sensing image, and meets the condition that the unchanged land type presents similar characteristics and the changed land type presents different characteristics as far as possible. The problem faced by the pixel-oriented change detection method is that the processing result is noisy and excessively dependent on the threshold selection.
Disclosure of Invention
The invention provides a remote sensing image change detection method and system based on a U-net improved algorithm, aims to solve the problems of low detection precision, poor anti-noise performance and the like of the existing change detection method, increases the network depth to improve the algorithm precision by applying a semantic segmentation network to the change detection direction and improving, and simultaneously extracts multi-scale context information by combining with ASPP (advanced application program protocol), thereby improving the remote sensing image change detection precision and the anti-noise capability and simultaneously improving the operation efficiency under the condition of ensuring the precision.
In order to solve the technical problem, the invention provides a remote sensing image change detection method based on a U-net improved algorithm, which comprises the following steps:
preprocessing the remote sensing images to be detected, and then integrating the preprocessed remote sensing images in different time phases to obtain an input image;
based on the improved U-net network, performing down-sampling coding and up-sampling decoding on the input image to complete analysis processing on the input image; the improved U-net network adopts a Resnet-101 network to extract data characteristics in an encoder part, and adds a cavity space pyramid pooling design ASPP between the encoder and the decoder for extracting data multi-scale context information;
and outputting a binary image of whether the remote sensing image to be detected changes or not based on the analysis processing result of the input image.
Further, the preprocessing the remote sensing image to be detected includes:
enhancing the remote sensing image to be detected, and processing the remote sensing images in different time phases by using histogram matching; wherein the enhancement operation includes flipping and zooming in or out.
Further, the integrating the preprocessed remote sensing images of different time phases includes:
splicing the preprocessed remote sensing images of two different time phases to form an input image of an 8+2 channel; wherein 8 channels are four-channel images of two time phases, and 2 channels are label channels whether to change or not.
Further, the performing down-sampling coding and up-sampling decoding on the input image based on the improved U-net network to complete the analysis processing on the input image includes:
utilizing a Resnet-101 network to perform downsampling feature extraction on the input image;
carrying out feature coding on feature information extracted by the Resnet-101 network by using ASPP;
and upsampling the coded data, and simultaneously connecting a decoder part for scale fusion.
Further, the Resnet-101 network includes a plurality of residual blocks, each of which includes three layers of convolutions.
Further, the ASPP includes one 1 × 1 convolution kernel and three 3 × 3 convolution kernels, and features at the image level.
Further, the decoder partial loss function in the improved U-net network is as follows:
wherein, LT is the probability of accurate label, PT is the probability of accurate prediction, w is the network weight, and the value range of the Loss function of Loss is [0,1 ].
Accordingly, in order to solve the above technical problem, the present invention further provides a remote sensing image change detection system based on a U-net improved algorithm, the remote sensing image change detection system based on the U-net improved algorithm comprising:
the image processing module is used for preprocessing the remote sensing images to be detected and then integrating the preprocessed remote sensing images in different time phases to obtain input images;
the analysis processing module is used for carrying out down-sampling coding and up-sampling decoding on the input image based on the improved U-net network so as to finish the analysis processing of the input image; the encoder of the improved U-net network adopts a Resnet-101 network to extract data characteristics, and a cavity space pyramid pooling design ASPP is added between the encoder and the decoder and used for extracting data multi-scale context information;
and the result output module is used for outputting a binary image of whether the remote sensing image to be detected changes or not based on the analysis processing result of the input image.
Further, the image processing module includes:
the image preprocessing unit is used for enhancing the remote sensing images to be detected and processing the remote sensing images in different time phases by using histogram matching; wherein the enhancement operation comprises flipping and zooming in or out;
the image integration unit is used for splicing the preprocessed remote sensing images in two different time phases to form an input image of an 8+2 channel; wherein 8 channels are four-channel images of two time phases, and 2 channels are label channels whether to change or not.
Further, the analysis processing module includes:
the Resnet-101 network unit is used for performing downsampling feature extraction on the input image by utilizing the Resnet-101 network;
the ASPP unit is used for carrying out feature coding on the feature information extracted by the Resnet-101 network unit by utilizing ASPP;
and the up-sampling decoding unit is used for up-sampling the coded data and simultaneously connecting the decoder part for scale fusion.
The technical scheme of the invention has the following beneficial effects:
the invention applies the network structure of the semantic segmentation direction to the field of change detection, and introduces a residual learning mechanism to ensure that an encoder can quickly converge and deepen the number of network layers, accelerate the characteristic extraction speed and solve the problems of gradient disappearance and gradient explosion; meanwhile, the perception capability of the network to the image features is enhanced by using the ASPP, and the efficiency is higher than that of other deep learning methods while the algorithm precision is ensured; the algorithm can be kept at a higher level in precision and efficiency, and has higher precision and stronger robustness. The method is suitable for the field of change detection of remote sensing images, can be popularized to other fields, and has important significance.
Drawings
FIG. 1 is a flow chart of the method for detecting changes in remote sensing images based on a U-net improved algorithm according to the present invention;
FIG. 2 is a network structure diagram of the method for detecting changes in remote sensing images according to the present invention;
fig. 3 is a network architecture diagram of a Resnet-101 network.
Detailed Description
In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments.
First embodiment
Referring to fig. 1 to 3, the present embodiment provides a remote sensing image change detection method based on a U-net improved algorithm, as shown in fig. 1, the remote sensing image change detection method includes:
s101, preprocessing remote sensing images to be detected, and integrating the preprocessed remote sensing images in different time phases to obtain an input image;
it should be noted that, the process of preprocessing the remote sensing image to be detected in this embodiment includes:
enhancing the remote sensing image to be detected, and processing the remote sensing images in different time phases by using histogram matching; wherein, the enhancing operation comprises operations such as turning over, enlarging or reducing.
The process of integrating the preprocessed remote sensing images in different time phases comprises the following steps:
splicing the preprocessed remote sensing images of two different time phases to form an input image of an 8+2 channel; wherein 8 channels are four-channel images of two time phases, and 2 channels are label channels whether to change or not. That is, the present embodiment integrates two images of different phases to form an eight-channel image as an input of the algorithm. And simultaneously, carrying out one-hot coding processing on the label to enable the label to become label data of two channels.
S102, based on the improved U-net network, performing down-sampling coding and up-sampling decoding on the input image to complete analysis processing on the input image;
it should be noted that, in the embodiment, in order to overcome the problems of simple network structure and shallow hierarchy of the current-stage stack denoising autoencoder, the network structure is deepened as much as possible while the efficiency is ensured, so that the change detection precision is improved; in the embodiment, the U-net network is improved, the U-net network framework can combine the advantages of a plurality of networks at the present stage, so that a network with reasonable depth and capable of extracting deep-level features is built, the U-net network framework is mainly divided into an encoder part and a decoder part, data feature extraction is carried out on the encoder part, and the decoder analyzes and processes the extracted features.
In this embodiment, a U-net network is improved, so that the U-net network is applicable to the problem of remote sensing image change detection, and the accuracy of remote sensing image change detection is effectively improved while the calculation efficiency is ensured, and the structure of the U-net network is shown in fig. 2.
Specifically, after analyzing the deficiencies of the existing method, to increase the accuracy, a deepened network layer is selected, and to avoid the problems of gradient loss and gradient explosion caused by the deepened network layer, the embodiment uses the Resnet-101 network to replace the encoder part in the U-net network for data feature extraction, the depth of the Resnet-101 network is reasonable, the idea of residual learning is introduced, the problems of gradient explosion or incapability of training and the like caused by the too deep network can be avoided, and the training of the neural network can be accelerated very quickly. The Resnet-101 network is composed of a plurality of residual blocks, the structure of which is shown in FIG. 3, and each residual block is composed of three layers of convolution. Inputting an image, completing convolution and ReLU operation, and obtaining a residual mapping defined as:
F(x,{Wi})=W2δ(W1xi) (1)
in equation (1), xiIs the input vector of the residual block, δ denotes the ReLU operation, WiIs a connection matrix of matching size. At the same time, a picture is input into a block concatenation to obtain an identity mapping Wsxi. Inputting the remaining mapping and identity mapping element addition and obtaining the remaining block definition as:
xi+1=F(x,{Wi})+Wsxi(2)
in equation (2), when xiWhen the size of F is not the same as that of W, W is usedsWhere F (x, { W)i} may represent multiple convolutional layers.
Meanwhile, considering that the neighborhood information of each pixel has great influence on change detection, the embodiment adds space pyramid pooling design (ASPP) of void convolution between the encoder and the decoder for extracting data multi-scale context information, wherein the ASPP can solve the problem of resolution reduction caused by DCNNs, does not increase new learning parameters, and reduces loss of detail information while expanding the receptive field. Wherein ASPP comprises one 1 x 1 convolution kernel and three 3 x 3 convolution kernels, as well as image-level features. The expansion coefficients of three of the 3 x 3 convolution kernels are 6, 12 and 18, respectively.
In this embodiment, the decoder partial loss function is as follows:
in equation (3), LT is the probability that the tag is accurate, PT is the probability that the prediction is accurate, and w is the network weight. The value range of the Loss function is [0,1 ].
The U-net improved network of the embodiment has proper depth at the encoder part, can accurately extract the characteristics of the input image, has strong anti-noise capability, and can avoid the situation that more noise information is detected by facing the change of the pixel to a great extent. And the decoder uses the hole space Pyramid Pooling (ASPP) for extracting the multi-scale context information. The rest part of the decoder adopts a U-net network with simple structure. The algorithm has stronger robustness, so the scheme of the embodiment not only can obtain better effect in change detection, but also has higher efficiency.
Based on the above, in this embodiment, the step S102 specifically includes the following steps:
utilizing a Resnet-101 network to perform downsampling feature extraction on the input image;
carrying out feature coding on feature information extracted by the Resnet-101 network by using ASPP;
the coded data is up-sampled and simultaneously connected with a decoder part for scale fusion; here the way the decoder is connected is a jump connection;
and S103, outputting a binary image of whether the remote sensing image to be detected changes or not based on the analysis processing result of the input image.
It should be noted that, in this embodiment, the output of the step S103 is specifically an image with a 0-1 label to indicate whether the remote sensing image to be detected changes, so as to complete the change detection of the remote sensing image.
In the embodiment, a network structure in a semantic segmentation direction is applied to the field of change detection, and a residual learning mechanism is introduced, so that an encoder can quickly converge and deepen the number of network layers, the feature extraction speed is increased, and the problems of gradient disappearance and gradient explosion are solved; meanwhile, the perception capability of the network to the image features is enhanced by using the ASPP, and the efficiency is higher than that of other deep learning methods while the algorithm precision is ensured; the algorithm can be kept at a higher level in precision and efficiency, and has higher precision and stronger robustness. The method is suitable for the field of change detection of remote sensing images, can be popularized to other fields, and has important significance.
Second embodiment
The embodiment provides a remote sensing image change detection system based on a U-net improved algorithm, which comprises:
the image processing module is used for preprocessing the remote sensing images to be detected and then integrating the preprocessed remote sensing images in different time phases to obtain input images;
the analysis processing module is used for carrying out down-sampling coding and up-sampling decoding on the input image based on the improved U-net network so as to finish the analysis processing of the input image; the encoder of the improved U-net network adopts the Resnet-101 network to extract data characteristics, and a cavity space pyramid pooling design ASPP is added between the encoder and the decoder and used for extracting data multi-scale context information;
and the result output module is used for outputting a binary image whether the remote sensing image to be detected changes or not based on the analysis processing result of the input image.
Further, the image processing module includes:
the image preprocessing unit is used for performing enhancement operations such as turning, amplification and reduction on the remote sensing image to be detected and processing the remote sensing images in different time phases by using histogram matching;
the image integration unit is used for splicing the preprocessed remote sensing images in two different time phases to form an input image of an 8+2 channel; wherein 8 channels are four-channel images of two time phases, and 2 channels are label channels whether to change or not.
Further, the analysis processing module includes:
the Resnet-101 network unit is used for performing downsampling feature extraction on the input image by utilizing the Resnet-101 network;
the ASPP unit is used for carrying out feature coding on the feature information extracted by the Resnet-101 network unit by utilizing the ASPP;
and the up-sampling decoding unit is used for up-sampling the coded data and simultaneously connecting the decoder part for scale fusion.
The remote sensing image change detection system based on the U-net improved algorithm of the embodiment corresponds to the remote sensing image change detection method based on the U-net improved algorithm of the first embodiment; the functions realized by each functional module in the remote sensing image change detection system based on the U-net improved algorithm correspond to each flow step in the remote sensing image change detection method based on the U-net improved algorithm in the first embodiment one by one; therefore, it is not described herein.
Furthermore, it should be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It should also be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.
The foregoing is merely a preferred embodiment of the invention and it should be noted that while the preferred embodiment of the invention has been described, it will be apparent to those skilled in the art that, once apprised of the basic inventive concepts of the present invention, numerous modifications and adaptations can be made without departing from the principles of the invention and are intended to be within the scope of the invention; that is, the appended claims are intended to be construed to include preferred embodiments and all such variations and modifications as fall within the scope of the embodiments of the invention.
Claims (10)
1. A remote sensing image change detection method based on a U-net improved algorithm is characterized by comprising the following steps:
preprocessing the remote sensing images to be detected, and then integrating the preprocessed remote sensing images in different time phases to obtain an input image;
based on the improved U-net network, performing down-sampling coding and up-sampling decoding on the input image to complete analysis processing on the input image; the improved U-net network adopts a Resnet-101 network to extract data characteristics in an encoder part, and adds a cavity space pyramid pooling design ASPP between the encoder and the decoder for extracting data multi-scale context information;
and outputting a binary image of whether the remote sensing image to be detected changes or not based on the analysis processing result of the input image.
2. The remote sensing image change detection method based on the U-net improved algorithm as claimed in claim 1, wherein the preprocessing of the remote sensing image to be detected comprises:
enhancing the remote sensing image to be detected, and processing the remote sensing images in different time phases by using histogram matching; wherein the enhancement operation includes flipping and zooming in or out.
3. The method for detecting changes in remote sensing images based on the U-net improved algorithm as claimed in claim 1, wherein the integrating the preprocessed remote sensing images in different time phases comprises:
splicing the preprocessed remote sensing images of two different time phases to form an input image of an 8+2 channel; wherein 8 channels are four-channel images of two time phases, and 2 channels are label channels whether to change or not.
4. The method for detecting changes in remotely sensed images based on U-net improved algorithm as claimed in claim 1, wherein said U-net network based on improvement performs down-sampling coding and up-sampling decoding on said input image to complete the analysis process of said input image, and comprises:
utilizing a Resnet-101 network to perform downsampling feature extraction on the input image;
carrying out feature coding on feature information extracted by the Resnet-101 network by using ASPP;
and upsampling the coded data, and simultaneously connecting a decoder part for scale fusion.
5. The method for remote sensing image change detection based on U-net improved algorithm of claim 4, wherein the Resnet-101 network comprises a plurality of residual blocks, and each residual block comprises three layers of convolution.
6. The method for remote sensing image change detection based on U-net improved algorithm of claim 4, wherein the ASPP comprises one 1 x 1 convolution kernel and three 3 x 3 convolution kernels, and features at image level.
7. The method for detecting changes in remote sensing images based on U-net improved algorithm as claimed in any of claims 1-6, wherein the partial loss function of the decoder in the improved U-net network is as follows:
wherein, LT is the probability of accurate label, PT is the probability of accurate prediction, w is the network weight, and the value range of the Loss function of Loss is [0,1 ].
8. A remote sensing image change detection system based on a U-net improved algorithm is characterized by comprising the following components:
the image processing module is used for preprocessing the remote sensing images to be detected and then integrating the preprocessed remote sensing images in different time phases to obtain input images;
the analysis processing module is used for carrying out down-sampling coding and up-sampling decoding on the input image based on the improved U-net network so as to finish the analysis processing of the input image; the encoder of the improved U-net network adopts a Resnet-101 network to extract data characteristics, and a cavity space pyramid pooling design ASPP is added between the encoder and the decoder and used for extracting data multi-scale context information;
and the result output module is used for outputting a binary image of whether the remote sensing image to be detected changes or not based on the analysis processing result of the input image.
9. The remote sensing image change detection system based on U-net improved algorithm of claim 8, characterized in that the image processing module comprises:
the image preprocessing unit is used for enhancing the remote sensing images to be detected and processing the remote sensing images in different time phases by using histogram matching; wherein the enhancement operation comprises flipping and zooming in or out;
the image integration unit is used for splicing the preprocessed remote sensing images in two different time phases to form an input image of an 8+2 channel; wherein 8 channels are four-channel images of two time phases, and 2 channels are label channels whether to change or not.
10. The remote sensing image change detection system based on U-net improved algorithm of claim 8, characterized in that the analysis processing module comprises:
the Resnet-101 network unit is used for performing downsampling feature extraction on the input image by utilizing the Resnet-101 network;
the ASPP unit is used for carrying out feature coding on the feature information extracted by the Resnet-101 network unit by utilizing ASPP;
and the up-sampling decoding unit is used for up-sampling the coded data and simultaneously connecting the decoder part for scale fusion.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911073929.6A CN111047551B (en) | 2019-11-06 | 2019-11-06 | Remote sensing image change detection method and system based on U-net improved algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911073929.6A CN111047551B (en) | 2019-11-06 | 2019-11-06 | Remote sensing image change detection method and system based on U-net improved algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111047551A true CN111047551A (en) | 2020-04-21 |
CN111047551B CN111047551B (en) | 2023-10-31 |
Family
ID=70231899
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911073929.6A Active CN111047551B (en) | 2019-11-06 | 2019-11-06 | Remote sensing image change detection method and system based on U-net improved algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111047551B (en) |
Cited By (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111563906A (en) * | 2020-05-07 | 2020-08-21 | 南开大学 | Knee joint magnetic resonance image automatic segmentation method based on deep convolutional neural network |
CN111611968A (en) * | 2020-05-29 | 2020-09-01 | 中国科学院西北生态环境资源研究院 | Processing method of remote sensing image and remote sensing image processing model |
CN111860233A (en) * | 2020-07-06 | 2020-10-30 | 中国科学院空天信息创新研究院 | SAR image complex building extraction method and system based on attention network selection |
CN111915571A (en) * | 2020-07-10 | 2020-11-10 | 云南电网有限责任公司带电作业分公司 | Image change detection method, device, storage medium and equipment fusing residual error network and U-Net network |
CN111932553A (en) * | 2020-07-27 | 2020-11-13 | 北京航空航天大学 | Remote sensing image semantic segmentation method based on area description self-attention mechanism |
CN111931684A (en) * | 2020-08-26 | 2020-11-13 | 北京建筑大学 | Weak and small target detection method based on video satellite data identification features |
CN111986099A (en) * | 2020-06-30 | 2020-11-24 | 武汉大学 | Tillage monitoring method and system based on convolutional neural network with residual error correction fused |
CN112070779A (en) * | 2020-08-04 | 2020-12-11 | 武汉大学 | Remote sensing image road segmentation method based on convolutional neural network weak supervised learning |
CN112164082A (en) * | 2020-10-09 | 2021-01-01 | 深圳市铱硙医疗科技有限公司 | Method for segmenting multi-modal MR brain image based on 3D convolutional neural network |
CN112381172A (en) * | 2020-11-28 | 2021-02-19 | 桂林电子科技大学 | InSAR interference image phase unwrapping method based on U-net |
CN112396594A (en) * | 2020-11-27 | 2021-02-23 | 广东电网有限责任公司肇庆供电局 | Change detection model acquisition method and device, change detection method, computer device and readable storage medium |
CN112668494A (en) * | 2020-12-31 | 2021-04-16 | 西安电子科技大学 | Small sample change detection method based on multi-scale feature extraction |
CN112733777A (en) * | 2021-01-18 | 2021-04-30 | 华侨大学 | Road extraction method, device, equipment and storage medium for remote sensing image |
CN112767413A (en) * | 2021-01-06 | 2021-05-07 | 武汉大学 | Remote sensing image depth semantic segmentation method integrating region communication and symbiotic knowledge constraints |
CN112990041A (en) * | 2021-03-25 | 2021-06-18 | 武汉大学 | Remote sensing image building extraction method based on improved U-net |
CN113033454A (en) * | 2021-04-07 | 2021-06-25 | 桂林电子科技大学 | Method for detecting building change in urban video camera |
CN113065551A (en) * | 2021-03-19 | 2021-07-02 | 杭州迪英加科技有限公司 | Method for performing image segmentation using a deep neural network model |
CN113096208A (en) * | 2021-03-16 | 2021-07-09 | 天津大学 | Reconstruction method of neural network magnetic resonance image based on double-domain alternating convolution |
CN113204608A (en) * | 2021-05-27 | 2021-08-03 | 广州大学 | Automatic map updating method, storage medium and system based on remote sensing image |
CN113628180A (en) * | 2021-07-30 | 2021-11-09 | 北京科技大学 | Semantic segmentation network-based remote sensing building detection method and system |
CN113688956A (en) * | 2021-10-26 | 2021-11-23 | 西南石油大学 | Sandstone slice segmentation and identification method based on depth feature fusion network |
CN113901877A (en) * | 2021-09-13 | 2022-01-07 | 广州市城市规划勘测设计研究院 | Method, device and equipment for detecting change of remote sensing image building and storage medium |
CN114549972A (en) * | 2022-01-17 | 2022-05-27 | 中国矿业大学(北京) | Strip mine stope extraction method, apparatus, device, medium, and program product |
CN115019186A (en) * | 2022-08-08 | 2022-09-06 | 中科星图测控技术(合肥)有限公司 | Algorithm and system for remote sensing change detection |
CN115205710A (en) * | 2022-09-16 | 2022-10-18 | 北京理工大学 | Double-time-phase remote sensing image change detection method combined with color correction |
CN116051519A (en) * | 2023-02-02 | 2023-05-02 | 广东国地规划科技股份有限公司 | Method, device, equipment and storage medium for detecting double-time-phase image building change |
CN116664599A (en) * | 2023-06-01 | 2023-08-29 | 云南大学 | Image steganalysis method based on steganography area prediction |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018214195A1 (en) * | 2017-05-25 | 2018-11-29 | 中国矿业大学 | Remote sensing imaging bridge detection method based on convolutional neural network |
CN109447994A (en) * | 2018-11-05 | 2019-03-08 | 陕西师范大学 | In conjunction with the remote sensing image segmentation method of complete residual error and Fusion Features |
CN109919206A (en) * | 2019-02-25 | 2019-06-21 | 武汉大学 | A kind of remote sensing image ground mulching classification method based on complete empty convolutional neural networks |
CN109934154A (en) * | 2019-03-08 | 2019-06-25 | 北京科技大学 | A kind of remote sensing image variation detection method and detection device |
CN109934200A (en) * | 2019-03-22 | 2019-06-25 | 南京信息工程大学 | A kind of RGB color remote sensing images cloud detection method of optic and system based on improvement M-Net |
-
2019
- 2019-11-06 CN CN201911073929.6A patent/CN111047551B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018214195A1 (en) * | 2017-05-25 | 2018-11-29 | 中国矿业大学 | Remote sensing imaging bridge detection method based on convolutional neural network |
CN109447994A (en) * | 2018-11-05 | 2019-03-08 | 陕西师范大学 | In conjunction with the remote sensing image segmentation method of complete residual error and Fusion Features |
CN109919206A (en) * | 2019-02-25 | 2019-06-21 | 武汉大学 | A kind of remote sensing image ground mulching classification method based on complete empty convolutional neural networks |
CN109934154A (en) * | 2019-03-08 | 2019-06-25 | 北京科技大学 | A kind of remote sensing image variation detection method and detection device |
CN109934200A (en) * | 2019-03-22 | 2019-06-25 | 南京信息工程大学 | A kind of RGB color remote sensing images cloud detection method of optic and system based on improvement M-Net |
Non-Patent Citations (1)
Title |
---|
YUHAO WANG 等: "Dense Semantic Labeling with Atrous Spatial Pyramid Pooling and Decoder for High-Resolution Remote Sensing Imagery", vol. 11, pages 1 - 18 * |
Cited By (38)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111563906A (en) * | 2020-05-07 | 2020-08-21 | 南开大学 | Knee joint magnetic resonance image automatic segmentation method based on deep convolutional neural network |
CN111611968A (en) * | 2020-05-29 | 2020-09-01 | 中国科学院西北生态环境资源研究院 | Processing method of remote sensing image and remote sensing image processing model |
CN111611968B (en) * | 2020-05-29 | 2022-02-01 | 中国科学院西北生态环境资源研究院 | Processing method of remote sensing image and remote sensing image processing model |
CN111986099A (en) * | 2020-06-30 | 2020-11-24 | 武汉大学 | Tillage monitoring method and system based on convolutional neural network with residual error correction fused |
CN111860233B (en) * | 2020-07-06 | 2021-05-18 | 中国科学院空天信息创新研究院 | SAR image complex building extraction method and system based on attention network selection |
CN111860233A (en) * | 2020-07-06 | 2020-10-30 | 中国科学院空天信息创新研究院 | SAR image complex building extraction method and system based on attention network selection |
CN111915571A (en) * | 2020-07-10 | 2020-11-10 | 云南电网有限责任公司带电作业分公司 | Image change detection method, device, storage medium and equipment fusing residual error network and U-Net network |
CN111932553A (en) * | 2020-07-27 | 2020-11-13 | 北京航空航天大学 | Remote sensing image semantic segmentation method based on area description self-attention mechanism |
CN112070779A (en) * | 2020-08-04 | 2020-12-11 | 武汉大学 | Remote sensing image road segmentation method based on convolutional neural network weak supervised learning |
CN111931684A (en) * | 2020-08-26 | 2020-11-13 | 北京建筑大学 | Weak and small target detection method based on video satellite data identification features |
CN111931684B (en) * | 2020-08-26 | 2021-04-06 | 北京建筑大学 | Weak and small target detection method based on video satellite data identification features |
CN112164082A (en) * | 2020-10-09 | 2021-01-01 | 深圳市铱硙医疗科技有限公司 | Method for segmenting multi-modal MR brain image based on 3D convolutional neural network |
CN112396594A (en) * | 2020-11-27 | 2021-02-23 | 广东电网有限责任公司肇庆供电局 | Change detection model acquisition method and device, change detection method, computer device and readable storage medium |
CN112396594B (en) * | 2020-11-27 | 2024-03-29 | 广东电网有限责任公司肇庆供电局 | Method and device for acquiring change detection model, change detection method, computer equipment and readable storage medium |
CN112381172A (en) * | 2020-11-28 | 2021-02-19 | 桂林电子科技大学 | InSAR interference image phase unwrapping method based on U-net |
CN112668494A (en) * | 2020-12-31 | 2021-04-16 | 西安电子科技大学 | Small sample change detection method based on multi-scale feature extraction |
CN112767413A (en) * | 2021-01-06 | 2021-05-07 | 武汉大学 | Remote sensing image depth semantic segmentation method integrating region communication and symbiotic knowledge constraints |
CN112733777A (en) * | 2021-01-18 | 2021-04-30 | 华侨大学 | Road extraction method, device, equipment and storage medium for remote sensing image |
CN112733777B (en) * | 2021-01-18 | 2023-12-05 | 华侨大学 | Road extraction method, device and equipment of remote sensing image and storage medium |
CN113096208A (en) * | 2021-03-16 | 2021-07-09 | 天津大学 | Reconstruction method of neural network magnetic resonance image based on double-domain alternating convolution |
CN113096208B (en) * | 2021-03-16 | 2022-11-18 | 天津大学 | Reconstruction method of neural network magnetic resonance image based on double-domain alternating convolution |
CN113065551A (en) * | 2021-03-19 | 2021-07-02 | 杭州迪英加科技有限公司 | Method for performing image segmentation using a deep neural network model |
CN113065551B (en) * | 2021-03-19 | 2023-08-08 | 杭州迪英加科技有限公司 | Method for performing image segmentation using deep neural network model |
CN112990041A (en) * | 2021-03-25 | 2021-06-18 | 武汉大学 | Remote sensing image building extraction method based on improved U-net |
CN113033454A (en) * | 2021-04-07 | 2021-06-25 | 桂林电子科技大学 | Method for detecting building change in urban video camera |
CN113204608A (en) * | 2021-05-27 | 2021-08-03 | 广州大学 | Automatic map updating method, storage medium and system based on remote sensing image |
CN113628180B (en) * | 2021-07-30 | 2023-10-27 | 北京科技大学 | Remote sensing building detection method and system based on semantic segmentation network |
CN113628180A (en) * | 2021-07-30 | 2021-11-09 | 北京科技大学 | Semantic segmentation network-based remote sensing building detection method and system |
CN113901877A (en) * | 2021-09-13 | 2022-01-07 | 广州市城市规划勘测设计研究院 | Method, device and equipment for detecting change of remote sensing image building and storage medium |
CN113688956A (en) * | 2021-10-26 | 2021-11-23 | 西南石油大学 | Sandstone slice segmentation and identification method based on depth feature fusion network |
CN114549972A (en) * | 2022-01-17 | 2022-05-27 | 中国矿业大学(北京) | Strip mine stope extraction method, apparatus, device, medium, and program product |
CN114549972B (en) * | 2022-01-17 | 2023-01-03 | 中国矿业大学(北京) | Strip mine stope extraction method, device, equipment and medium |
CN115019186A (en) * | 2022-08-08 | 2022-09-06 | 中科星图测控技术(合肥)有限公司 | Algorithm and system for remote sensing change detection |
CN115205710A (en) * | 2022-09-16 | 2022-10-18 | 北京理工大学 | Double-time-phase remote sensing image change detection method combined with color correction |
CN116051519B (en) * | 2023-02-02 | 2023-08-22 | 广东国地规划科技股份有限公司 | Method, device, equipment and storage medium for detecting double-time-phase image building change |
CN116051519A (en) * | 2023-02-02 | 2023-05-02 | 广东国地规划科技股份有限公司 | Method, device, equipment and storage medium for detecting double-time-phase image building change |
CN116664599A (en) * | 2023-06-01 | 2023-08-29 | 云南大学 | Image steganalysis method based on steganography area prediction |
CN116664599B (en) * | 2023-06-01 | 2024-06-11 | 云南大学 | Image steganalysis method based on steganography area prediction |
Also Published As
Publication number | Publication date |
---|---|
CN111047551B (en) | 2023-10-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111047551B (en) | Remote sensing image change detection method and system based on U-net improved algorithm | |
CN112541503B (en) | Real-time semantic segmentation method based on context attention mechanism and information fusion | |
CN111461114B (en) | Multi-scale feature pyramid text detection method based on segmentation | |
CN108509978B (en) | Multi-class target detection method and model based on CNN (CNN) multi-level feature fusion | |
CN112966684B (en) | Cooperative learning character recognition method under attention mechanism | |
CN111210443A (en) | Deformable convolution mixing task cascading semantic segmentation method based on embedding balance | |
CN113505792B (en) | Multi-scale semantic segmentation method and model for unbalanced remote sensing image | |
CN109784183B (en) | Video saliency target detection method based on cascade convolution network and optical flow | |
CN109657551B (en) | Face detection method based on context information enhancement | |
CN111882620B (en) | Road drivable area segmentation method based on multi-scale information | |
CN113487618B (en) | Portrait segmentation method, portrait segmentation device, electronic equipment and storage medium | |
CN110059768A (en) | The semantic segmentation method and system of the merging point and provincial characteristics that understand for streetscape | |
CN111353544B (en) | Improved Mixed Pooling-YOLOV 3-based target detection method | |
CN116665176B (en) | Multi-task network road target detection method for vehicle automatic driving | |
CN113255837A (en) | Improved CenterNet network-based target detection method in industrial environment | |
CN117765361B (en) | Method for detecting building change area in double-time-phase remote sensing image based on residual neural network | |
CN112270366A (en) | Micro target detection method based on self-adaptive multi-feature fusion | |
CN114926734B (en) | Solid waste detection device and method based on feature aggregation and attention fusion | |
CN114120148B (en) | Method for detecting changing area of remote sensing image building | |
CN117197462A (en) | Lightweight foundation cloud segmentation method and system based on multi-scale feature fusion and alignment | |
CN112818818B (en) | Novel ultra-high-definition remote sensing image change detection method based on AFFPN | |
CN118247686A (en) | Training method, recognition method, system, equipment and medium for image detection model | |
CN113096133A (en) | Method for constructing semantic segmentation network based on attention mechanism | |
CN115082778B (en) | Multi-branch learning-based homestead identification method and system | |
CN116246172A (en) | Road extraction method and system based on road segmentation and intersection detection |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |