WO2023000159A1 - 高分辨率遥感影像半监督分类方法、装置、设备及介质 - Google Patents

高分辨率遥感影像半监督分类方法、装置、设备及介质 Download PDF

Info

Publication number
WO2023000159A1
WO2023000159A1 PCT/CN2021/107289 CN2021107289W WO2023000159A1 WO 2023000159 A1 WO2023000159 A1 WO 2023000159A1 CN 2021107289 W CN2021107289 W CN 2021107289W WO 2023000159 A1 WO2023000159 A1 WO 2023000159A1
Authority
WO
WIPO (PCT)
Prior art keywords
remote sensing
classification
model
segmentation model
sensing images
Prior art date
Application number
PCT/CN2021/107289
Other languages
English (en)
French (fr)
Inventor
刘康
安源
朱济帅
邓美环
王诒琬
Original Assignee
海南长光卫星信息技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 海南长光卫星信息技术有限公司 filed Critical 海南长光卫星信息技术有限公司
Priority to PCT/CN2021/107289 priority Critical patent/WO2023000159A1/zh
Publication of WO2023000159A1 publication Critical patent/WO2023000159A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology

Definitions

  • the invention relates to the field of classification of remote sensing images, in particular to a method, device, equipment and medium for semi-supervised classification of high-resolution remote sensing images.
  • remote sensing satellites As the most important part of remote sensing satellite products, high-resolution remote sensing images are widely used in agricultural production estimation, agricultural risk assessment, mineral investigation, land and resources investigation and other fields. In recent years, high-resolution remote sensing satellites have been launched rapidly, and image data sources have increased dramatically, providing abundant data resources for later applications.
  • the processing of remote sensing data generally includes pre-processing and advanced analysis. In post-application, image classification is the basic research to understand the coverage of ground objects.
  • the resolution of high-resolution remote sensing images is better than 1 meter (ie > 1 meter), which can clearly distinguish vegetation, water bodies, buildings and other ground objects.
  • Using high-resolution remote sensing images to classify ground features can grasp the details of ground features and understand the types of ground features.
  • There are many methods for remote sensing image classification which can be divided into supervised classification, unsupervised classification and semi-supervised classification from the perspective of using training samples.
  • Most conventional supervised classification methods start from the spectral characteristics of remote sensing images, and consider the spectral differences between different object types to distinguish pixels.
  • there is not much spectral information and usually only includes four bands of RGB and near-infrared, which cannot contain rich spectral information. After classification using spectral classification methods, the classification accuracy is low.
  • the present invention is proposed to provide a semi-supervised classification method, device, equipment and medium for high-resolution remote sensing images that overcome the above problems or at least partially solve the above problems.
  • a semi-supervised classification method for high-resolution remote sensing images comprising:
  • the classification model is used to classify the remote sensing images to be classified.
  • the preprocessing of the remote sensing images includes:
  • the preparation of a ground object classification sample set according to the processed remote sensing images includes:
  • the processed remote sensing image is sliced, and the marked grid corresponding to the position is cut out at the same time, so as to obtain a group of pictures and labels of the same size as a sample set of ground object classification.
  • the Unet++ network adopts the method of multi-level upsampling and layer-skip connection to extract multi-layer features.
  • the Unet++ network includes a downsampling layer, an upsampling layer, and an intermediate layer for feature extraction of the downsampling layer ;
  • the downsampling layer is added with the feature extraction part of the EfficientB4 model.
  • the construction of a threshold segmentation model based on the near-infrared band includes:
  • the ground feature classification sample set is matched with the near-infrared band threshold histogram, and the threshold area of water body and vegetation in the near-infrared band threshold histogram is selected to construct a threshold segmentation model.
  • the model fusion of the semantic segmentation model and the threshold segmentation model to obtain a classification model includes:
  • the classification result determined to be correct is used as a new object classification sample set, and the semantic segmentation model is continuously trained by using the transfer learning method to obtain a classification model.
  • the embodiment of the present invention also provides a semi-supervised classification device for high-resolution remote sensing images, including:
  • the image processing module is used for preprocessing the remote sensing images
  • the sample set making module is used to make a sample set of ground object classification according to the processed remote sensing image
  • the first model building module is used to build a remote sensing image semantic segmentation model based on the Unet++ network, and trains the semantic segmentation model through the feature classification sample set;
  • the second model building block is used to build a threshold segmentation model based on the near-infrared band
  • a model fusion module configured to perform model fusion on the semantic segmentation model and the threshold segmentation model to obtain a classification model
  • An image classification module configured to use the classification model to classify remote sensing images to be classified.
  • the embodiment of the present invention also provides a semi-supervised classification device for high-resolution remote sensing images, including a processor and a memory, wherein, when the processor executes the computer program stored in the memory, the above-mentioned A semi-supervised classification method for high-resolution remote sensing images.
  • An embodiment of the present invention also provides a computer-readable storage medium for storing a computer program, wherein, when the computer program is executed by a processor, the above-mentioned method for semi-supervised classification of high-resolution remote sensing images as provided in the embodiment of the present invention is implemented .
  • a semi-supervised classification method for high-resolution remote sensing images includes: preprocessing the remote sensing images; making a classification sample set based on the processed remote sensing images; The remote sensing image semantic segmentation model of the Unet++ network, and the semantic segmentation model is trained through the object classification sample set; the threshold segmentation model based on the near-infrared band is constructed; the semantic segmentation model and the threshold segmentation model are model-fused to obtain the classification model; use The classification model classifies the remote sensing images to be classified.
  • the present invention builds a remote sensing image semantic segmentation model based on the Unet++ network and a threshold segmentation model based on the near-infrared band, and then uses a multi-model fusion method to fuse the texture information of the remote sensing image with the spectral information of the near-infrared band, and then perform high-resolution High-rate remote sensing image classification can improve classification accuracy.
  • the present invention also provides corresponding devices, equipment, and computer-readable storage media for the semi-supervised classification method of high-resolution remote sensing images, which further makes the above method more practical.
  • the device, equipment, and computer-readable storage media have corresponding The advantages.
  • Fig. 1 shows the flow chart of the semi-supervised classification method of high-resolution remote sensing image provided by the embodiment of the present invention
  • FIG. 2 shows a schematic diagram of a semi-supervised classification method for high-resolution remote sensing images provided by an embodiment of the present invention
  • Fig. 3 shows the structural representation of the Unet++ network that the embodiment of the present invention provides
  • Fig. 4 shows the schematic structural diagram of the convolution block in the Unet++ network provided by the embodiment of the present invention
  • Fig. 5 shows a schematic structural diagram of the residual block in the Unet++ network provided by the embodiment of the present invention
  • Fig. 6 shows the near-infrared band threshold histogram provided by the embodiment of the present invention
  • FIG. 7 shows a display diagram of classification results provided by an embodiment of the present invention.
  • Fig. 8 shows a schematic structural diagram of an apparatus for semi-supervised classification of high-resolution remote sensing images provided by an embodiment of the present invention.
  • the present invention provides a semi-supervised classification method for high-resolution remote sensing images, as shown in Figure 1 and Figure 2, comprising the following steps:
  • the remote sensing image semantic segmentation model based on the Unet++ network and a threshold segmentation model based on the near-infrared band, and then using a multi-model fusion method, the remote sensing
  • the texture information of the image and the spectral information of the near-infrared band are fused, and then the classification of high-resolution remote sensing images can improve the classification accuracy.
  • step S101 performs preprocessing on remote sensing images, which may specifically include: performing panchromatic and multispectral image processing on remote sensing images Fusion, radiation correction, atmospheric correction, geometric correction and other processing to obtain a high-resolution remote sensing image with four bands (RGB and near-infrared).
  • Step S102 is based on the processed remote sensing images.
  • Object classification sample set which may specifically include:
  • the processed remote sensing image is sliced; specifically, the remote sensing image is cut into small slices for training.
  • the slice size can be set to 512 ⁇ 512, and the slices do not overlap; at the same time, the label of the corresponding position is cut out.
  • a set of images and labels of the same size is obtained as a sample set for object classification.
  • the Unet++ network adopts the method of multi-level upsampling and layer-skip connection to extract multi-layer features. It should be noted that the Unet network is a commonly used segmentation network model in semantic segmentation.
  • downsampling is performed through convolution, layer after layer features are extracted, and then upsampling is performed, and the difference between downsampling and upsampling is
  • the feature connection of the model because the shape of this model structure diagram is similar to U-shaped, so it is named Unet, where the process of downsampling is an encoding process, and upsampling is a process of decoding; while the Unet++ network is based on Unet.
  • the method of multi-level upsampling and layer-skip connection is adopted to extract features of more layers.
  • the Unet++ network can specifically include a downsampling layer, an upsampling layer, and an intermediate layer for feature extraction of the downsampling layer ;
  • the feature extraction part of the EfficientB4 model is added to the downsampling layer. That is to say, on the basis of Unet, the present invention adds the feature extraction part of the EfficientB4 model into the encoding process of the Unet++ network, improves the network structure, and extracts more features.
  • step S103 builds a remote sensing image semantic segmentation model based on the Unet++ network, which may specifically include the following steps:
  • a convolution block includes a convolution layer, BatchNormalization (BN) layer, and a LeakyRelU activation function;
  • the downsampling layer is the part of feature extraction. It is the same as the Unet network.
  • four downsampling layers are used as the downsampling layer of the Unet++ network.
  • the downsampling layer of the network is obtained from EfficientB4 , that is, the 342nd, 154th, 92nd, and 30th layers of EfficientB4 are used as the four layers conv4, conv3, conv2, and conv1 of the downsampling in Unet++;
  • the middle layer is a further feature extraction for the downsampling layer, and the features of different levels are not given.
  • conv4 it is coded as deconv4, and then extracted three times to obtain the three-level feature layer deconv4_up1, deconv4_up2, deconv4_up3, then extract the features of conv4 as deconv3, and extract the features deconv3_up1, deconv3_up2, add deconv3, conv3 and deconv4_up1 to obtain uconv3, the Uconv3 is encoded as deconv2, and the feature deconv2_up1 is extracted, and then deconv2, conv2, deconv4_up2, deconv3_up1 are added to obtain uconv2, uconv2 is encoded as deconv1, and then conv1, deconv1, deconv2_up1, deconv3_up2, deconv4_up3 are added to obtain uconv1, and encoded It is uconv0, and finally make a convolution of uconv0, reduce the feature to 1 dimension,
  • the loss function using the dice loss between the network prediction result and the real label as the loss function.
  • step S104 constructs a threshold segmentation model based on the near-infrared band, which may specifically include: using the threshold segmentation method to obtain processed remote sensing images The threshold histogram of the near-infrared band; match the object classification sample set with the threshold histogram of the near-infrared band, and select the threshold area of water and vegetation in the threshold histogram of the near-infrared band to build a threshold segmentation model.
  • band 4 is used to judge water bodies and vegetation.
  • the method of threshold segmentation is used to extract only band 4 in the high-resolution remote sensing image, and display the histogram.
  • the abscissa of the histogram ranges from 0 to 255, as shown in Figure 6.
  • the training samples and The histogram is matched, and a section of the water body in the histogram is selected as the reflectance part of the water body in the image, usually on the left, to record the threshold area of the water body.
  • the threshold area of the vegetation is selected.
  • the data to be classified is within the threshold of water body or vegetation, it is marked as water body or vegetation. In this way, making full use of the characteristics of the near-infrared band can increase the classification accuracy of specific targets to a certain extent.
  • step S105 performs model fusion of the semantic segmentation model and the threshold segmentation model to obtain the classification model, which may specifically include: when the semantic segmentation model When the output result of the threshold segmentation model and the output result of the threshold segmentation model are both water bodies or vegetation, or when the output result of the semantic segmentation model is bare soil or impermeable surface and the output result of the threshold segmentation model is other, it is determined that the classification result is correct; The classification result judged to be correct is used as a new object classification sample set, and the transfer learning method is used to continue training the semantic segmentation model to obtain the classification model.
  • semantic segmentation is used to classify remote sensing images, and at the same time, for some types of ground objects such as water bodies, vegetation, etc.
  • the near-infrared band features are used to extract, and the two results are fused at the decision level, and then the semi-supervised classification method is used to increase samples, and Classification again can obtain higher classification accuracy.
  • step S105 uses a decision-level fusion method.
  • the output result of the semantic segmentation model is a two-dimensional matrix, and each position represents the category of the pixel, including four categories in total.
  • the result of threshold segmentation is also a two-dimensional matrix. Each location represents the category of the pixel. Unlike semantic segmentation, it only contains three categories, water body, vegetation, and others. For each pixel, when the result of semantic segmentation is bare soil or impermeable surface, the result of threshold segmentation is other, or both are water bodies or vegetation, the classification is considered correct. When the classification results are different, this part is not classified, and the classified one is used as a new training sample, and the transfer learning method is used to continue training the semantic segmentation model to obtain the final classification result.
  • step S106 uses the classification model to classify the remote sensing images to be classified, which may specifically include: first, slice the remote sensing images to be classified, slice The size is the same as step 1, and then each slice is input into the classification model for classification, after expansion, corrosion and other post-processing operations, and then the classification results of the last slices are stitched together, coordinate information is added, and a tiff file is generated, an example is shown in Figure 7 shown.
  • the above-mentioned semi-supervised classification method for high-resolution remote sensing images mainly considers the texture features of high-resolution remote sensing images and the physical characteristics of the near-infrared band, and uses the improved Unet++ semantic segmentation network and threshold segmentation In a combined way, the semi-supervised classification method is used to expand the training samples, and the semantic segmentation model is trained again to obtain higher classification accuracy, and finally the classification results are generated after classification.
  • an embodiment of the present invention also provides a semi-supervised classification device for high-resolution remote sensing images. Since the problem-solving principle of the device is similar to the aforementioned semi-supervised classification method for high-resolution remote sensing images, the device's For the implementation, please refer to the implementation of the semi-supervised classification method for high-resolution remote sensing images, and the repetition will not be repeated.
  • the high-resolution remote sensing image semi-supervised classification device provided by the embodiment of the present invention, as shown in Figure 8, specifically includes:
  • An image processing module configured to preprocess remote sensing images
  • a sample set making module 12 configured to make a sample set for classification of features according to the processed remote sensing images
  • the first model building module 13 is used to build a remote sensing image semantic segmentation model based on the Unet++ network, and trains the semantic segmentation model through the feature classification sample set;
  • the second model construction module 14 is used to construct a threshold segmentation model based on the near-infrared band
  • the model fusion module 15 is used to carry out model fusion with the semantic segmentation model and the threshold segmentation model to obtain the classification model;
  • the image classification module 16 is configured to use a classification model to classify the remote sensing image to be classified.
  • the texture information of the remote sensing images and the spectral information in the near-infrared band can be fused through the interaction of the above six modules, and then the high-resolution Remote sensing image classification improves classification accuracy.
  • the embodiment of the present invention also discloses a semi-supervised classification device for high-resolution remote sensing images, including a processor and a memory; wherein, when the processor executes the computer program stored in the memory, the high-resolution remote sensing image disclosed in the foregoing embodiments is realized.
  • Image semi-supervised classification method including a processor and a memory; wherein, when the processor executes the computer program stored in the memory, the high-resolution remote sensing image disclosed in the foregoing embodiments is realized.
  • the present invention also discloses a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, the aforementioned semi-supervised classification method for high-resolution remote sensing images is realized.
  • RAM random access memory
  • ROM read-only memory
  • EEPROM electrically programmable ROM
  • EEPROM electrically erasable programmable ROM
  • registers hard disk, removable disk, CD-ROM, or any other Any other known storage medium.
  • a semi-supervised classification method for high-resolution remote sensing images includes: preprocessing the remote sensing images; making a sample set of ground object classification according to the processed remote sensing images; constructing a remote sensing image based on the Unet++ network Image semantic segmentation model, and train the semantic segmentation model through the object classification sample set; build a threshold segmentation model based on the near-infrared band; fuse the semantic segmentation model and the threshold segmentation model to obtain a classification model; use the classification model to treat classification classification of remote sensing images.
  • the present invention also provides corresponding devices, equipment, and computer-readable storage media for the semi-supervised classification method of high-resolution remote sensing images, which further makes the above method more practical.
  • the device, equipment, and computer-readable storage media have corresponding The advantages.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Image Processing (AREA)

Abstract

一种高分辨率遥感影像半监督分类方法、装置、设备及介质,对遥感影像进行预处理;根据处理后的遥感影像,制作地物分类样本集;构建基于Unet++网络的遥感影像语义分割模型,并通过地物分类样本集对语义分割模型进行训练;构建基于近红外波段的阈值分割模型;将语义分割模型和阈值分割模型进行模型融合,获取分类模型;使用分类模型对待分类遥感影像进行分类。这样通过构建基于Unet++网络的遥感影像语义分割模型和基于近红外波段的阈值分割模型,然后使用多模型融合的方法,使遥感影像的纹理信息和近红外波段的光谱信息融合,再对高分辨率遥感影像分类,能够提高分类精度。

Description

高分辨率遥感影像半监督分类方法、装置、设备及介质 技术领域
本发明涉及遥感影像分类领域,具体涉及一种高分辨率遥感影像半监督分类方法、装置、设备及介质。
背景技术
高分辨率遥感影像作为遥感卫星产品中最主要的部分,广泛应用于农业估产、农业风险评估、矿产调查、国土资源调查等领域。近些年以来,高分辨率遥感卫星发射迅猛,影像数据源暴增,为后期应用提供了丰富的数据资源。遥感数据的处理一般包括前期处理和高级分析,后期应用中,影像分类是了解地物覆盖的基础研究。
一般来说,高分辨率遥感影像的分辨率优于1米(即>1米),可以较为清晰地分辨植被、水体、建筑等地物目标。使用高分辨率遥感影像做地物分类,能够把握地物细节,了解地物类型。遥感影像分类的方法有很多,从使用训练样本的角度,分为监督分类、非监督分类、半监督分类。比较常规的监督分类的方法大多从遥感影像的光谱特征出发,考虑不同地物类型之间的光谱差异来区分像元。而对于高分辨率遥感影像,光谱信息不多,通常只包含RGB和近红外四个波段,无法包含丰富的光谱信息,使用光谱分类的方法进行分类后,分类精度较低。
因此,如何解决高分辨率遥感影像分类精度低的问题,是本领域技术人员亟待解决的技术问题。
发明内容
鉴于上述问题,提出了本发明以便提供一种克服上述问题或者至少部分地解决上述问题的一种高分辨率遥感影像半监督分类方法、装置、设备及介质。
一种高分辨率遥感影像半监督分类方法,包括:
对遥感影像进行预处理;
根据处理后的所述遥感影像,制作地物分类样本集;
构建基于Unet++网络的遥感影像语义分割模型,并通过所述地物分类样本集对所述语义分割模型进行训练;
构建基于近红外波段的阈值分割模型;
将所述语义分割模型和所述阈值分割模型进行模型融合,获取分类模型;
使用所述分类模型对待分类遥感影像进行分类。
优选地,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,所述对遥感影像进行预处理,包括:
对遥感影像进行全色和多光谱影像融合、辐射校正、大气校正、几何校正的处理。
优选地,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,所述根据处理后的所述遥感影像,制作地物分类样本集,包括:
建立与处理后的所述遥感影像相同大小的矢量,并将所述矢量全要素分割,分成水体、植被、裸土、不透水面四个类型;
对分成的类型进行标注,并根据类型字段不同,转换为栅格数据;
对处理后的所述遥感影像进行切片处理,同时裁剪出对应位置的标注的栅格,得到一组大小相同的图片和标签作为地物分类样本集。
优选地,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,所述Unet++网络采用多级上采样和跳层连接的方法,提取多层特征。
优选地,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,所述Unet++网络包括下采样层,上采样层,以及用于对所述下采样层进行特征抽取的中间层;其中,
所述下采样层添加有EfficientB4模型的特征提取部分。
优选地,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,所述构建基于近红外波段的阈值分割模型,包括:
使用阈值分割法获取处理后的所述遥感影像的近红外波段阈值直方图;
将所述地物分类样本集和所述近红外波段阈值直方图进行匹配,选取 出所述近红外波段阈值直方图中水体和植被的阈值区域,以构建阈值分割模型。
优选地,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,所述将所述语义分割模型和所述阈值分割模型进行模型融合,获取分类模型,包括:
当所述语义分割模型的输出结果与所述阈值分割模型的输出结果均为水体或植被时,或,当所述语义分割模型的输出结果为裸土或不透水面且所述阈值分割模型的输出结果为其他时,判定分类结果正确;
将判定为正确的分类结果作为新的地物分类样本集,使用迁移学习的方法继续训练所述语义分割模型,获取分类模型。
本发明实施例还提供了一种高分辨率遥感影像半监督分类装置,包括:
影像处理模块,用于对遥感影像进行预处理;
样本集制作模块,用于根据处理后的所述遥感影像,制作地物分类样本集;
第一模型构建模块,用于构建基于Unet++网络的遥感影像语义分割模型,并通过所述地物分类样本集对所述语义分割模型进行训练;
第二模型构建模块,用于构建基于近红外波段的阈值分割模型;
模型融合模块,用于将所述语义分割模型和所述阈值分割模型进行模型融合,获取分类模型;
影像分类模块,用于使用所述分类模型对待分类遥感影像进行分类。
本发明实施例还提供了一种高分辨率遥感影像半监督分类设备,包括处理器和存储器,其中,所述处理器执行所述存储器中存储的计算机程序时实现如本发明实施例提供的上述高分辨率遥感影像半监督分类方法。
本发明实施例还提供了一种计算机可读存储介质,用于存储计算机程序,其中,所述计算机程序被处理器执行时实现如本发明实施例提供的上述高分辨率遥感影像半监督分类方法。
从上述技术方案可以看出,本发明所提供的一种高分辨率遥感影像半监督分类方法,包括:对遥感影像进行预处理;根据处理后的遥感影像, 制作地物分类样本集;构建基于Unet++网络的遥感影像语义分割模型,并通过地物分类样本集对语义分割模型进行训练;构建基于近红外波段的阈值分割模型;将语义分割模型和阈值分割模型进行模型融合,获取分类模型;使用分类模型对待分类遥感影像进行分类。
本发明通过构建基于Unet++网络的遥感影像语义分割模型和基于近红外波段的阈值分割模型,然后使用多模型融合的方法,使遥感影像的纹理信息和近红外波段的光谱信息融合,再对高分辨率遥感影像分类,能够提高分类精度。
此外,本发明还针对高分辨率遥感影像半监督分类方法提供了相应的装置、设备及计算机可读存储介质,进一步使得上述方法更具有实用性,该装置、设备及计算机可读存储介质具有相应的优点。
上述说明仅是本发明技术方案的概述,为了能够更清楚了解本发明的技术手段,而可依照说明书的内容予以实施,并且为了让本发明的上述和其它目的、特征和优点能够更明显易懂,以下特举本发明的具体实施方式。
附图说明
通过阅读下文优选实施方式的详细描述,各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的,而并不认为是对本发明的限制。而且在整个附图中,用相同的参考符号表示相同的部件。在附图中:
图1示出了本发明实施例提供的高分辨率遥感影像半监督分类方法的流程图;
图2示出了本发明实施例提供的高分辨率遥感影像半监督分类方法的示意图;
图3示出了本发明实施例提供的Unet++网络的结构示意图;
图4示出了本发明实施例提供的Unet++网络中卷积block的结构示意图;
图5示出了本发明实施例提供的Unet++网络中残差block的结构示意图;
图6示出了本发明实施例提供的近红外波段阈值直方图;
图7示出了本发明实施例提供的分类结果展示图;
图8示出了本发明实施例提供的高分辨率遥感影像半监督分类装置的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
本发明提供一种高分辨率遥感影像半监督分类方法,如图1和图2所示,包括以下步骤:
S101、对遥感影像进行预处理;
S102、根据处理后的遥感影像,制作地物分类样本集;
S103、构建基于Unet++网络的遥感影像语义分割模型,并通过地物分类样本集对语义分割模型进行训练;
S104、构建基于近红外波段的阈值分割模型;
S105、将语义分割模型和阈值分割模型进行模型融合,获取分类模型;
S106、使用分类模型对待分类遥感影像进行分类。
在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,通过构建基于Unet++网络的遥感影像语义分割模型和基于近红外波段的阈值分割模型,然后使用多模型融合的方法,使遥感影像的纹理信息和近红外波段的光谱信息融合,再对高分辨率遥感影像分类,能够提高分类精度。
进一步地,在具体实施时,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,步骤S101对遥感影像进行预处理,具体可以包括:对遥感影像进行全色和多光谱影像融合、辐射校正、大气校正、几何校正等处理,获取一个具有四波段(RGB和近红外)的高分辨率遥感影像。
在具体实施时,在本发明实施例提供的上述高分辨率遥感影像半监督 分类方法中,由于遥感影像尺寸过大,需要裁剪为切片,赋予属性,步骤S102根据处理后的遥感影像,制作地物分类样本集,具体可以包括:
首先,建立与处理后的遥感影像相同大小的矢量,并将矢量全要素分割,分成水体、植被、裸土、不透水面四个类型;
然后,对分成的类型进行标注,并根据类型字段不同,转换为栅格数据;具体可以使用shpfile文件进行标注,然后将shpfile根据类型字段不同,转换为栅格数据;
最后,对处理后的遥感影像进行切片处理;具体地,将遥感影像切成一个个小的切片便于训练,切片大小可以设置为512×512,切片之间不重叠;同时裁剪出对应位置的标注的栅格,得到一组大小相同的图片和标签作为地物分类样本集。
在具体实施时,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,如图3所示,Unet++网络采用多级上采样和跳层连接的方法,提取多层特征。需要说明的是,Unet网络是语义分割中常用的一种分割网络模型,首先通过卷积进行下采样,提取一层又一层的特征,然后进行上采样,并将下采样和上采样之间的特征连接,由于这个模型结构图形状类似U形,所以名为Unet,其中下采样的过程是一个编码的过程,上采样是解码的过程;而Unet++网络是在Unet的基础上进行了增加,采用了多级上采样和跳层连接的方法,提取更多层的特征。
在具体实施时,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,Unet++网络具体可以包括下采样层,上采样层,以及用于对下采样层进行特征抽取的中间层;其中,下采样层添加有EfficientB4模型的特征提取部分。也就是说,本发明在Unet的基础上,将EfficientB4模型的特征提取部分加入到Unet++网络的编码过程中,改进了网络结构,提取了更多的特征。
具体地,步骤S103构建基于Unet++网络的遥感影像语义分割模型,具体可以包括以下步骤:
首先,定义一个卷积block,如图4所示,一个卷积block包括一个卷积层、BatchNormalization(BN)层、一个LeakyRelU激活函数;
然后,定义一个残差block,如图5所示,将残差block网络的输入经过LeakReLU和BN层之后,再经过上面定义的两个卷积block,之后加上原始输入经过BN层处理之后的结果;
之后,建立Unet++网络的下采样层、上采用层和中间层。下采样层是特征提取的部分,与Unet网络相同,这里使用四个下采样层作为Unet++网络的下采样层,不同的是,为了更深入地提出特征,该网络的下采样层是从EfficientB4获取,即分别从EfficientB4的第342、154、92、30层作为Unet++中下采样的四个层conv4、conv3、conv2、conv1;中间层是对下采样层的进一步特征抽取,不予不同层级的特征,抽取次数也不一样。对于conv4,编码为deconv4,再抽取三次,获得三级特征层deconv4_up1、deconv4_up2、deconv4_up3,然后将conv4提取特征为deconv3,并提取特征deconv3_up1,deconv3_up2,将deconv3、conv3与deconv4_up1相加,获得uconv3,将uconv3编码为deconv2,并提取特征deconv2_up1,然后将deconv2、conv2、deconv4_up2、deconv3_up1相加,得到uconv2,将uconv2编码为deconv1,然后将conv1、deconv1、deconv2_up1、deconv3_up2、deconv4_up3相加,获得uconv1,并编码为uconv0,最后将uconv0做一个卷积,将特征降到1维,作为网络输出;
最后,定义损失函数,使用网络预测结果与真实标签之间的dice loss作为损失函数。使用余弦退火的方法设定learning rate(lr),初始lr设置为0.001,epoch设置为30,batchsize设置为32,然后开始训练模型。
在具体实施时,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,步骤S104构建基于近红外波段的阈值分割模型,具体可以包括:使用阈值分割法获取处理后的遥感影像的近红外波段阈值直方图;将地物分类样本集和近红外波段阈值直方图进行匹配,选取出近红外波段阈值直方图中水体和植被的阈值区域,以构建阈值分割模型。
具体地,考虑到近红外波段对水体的吸收较强,而且植被中含有较多的水体,所以水体和植被的反射率较低,使用波段四来判断水体和植被。步骤S104中使用阈值分割的方法,只提取高分辨率遥感影像中的波段四,显示直方图,对于8位影像,直方图横坐标范围为0到255,如图6所示, 将训练样本和直方图进行匹配,选取直方图中水体聚集的一段作为影像中水体的反射率部分,通常靠左,记录水体的阈值区域,同样,选取植被的阈值区域。待分类的数据在水体或者植被的阈值内时,标记为水体或植被。这样充分利用近红外波段的特征,能够一定程度上增加特定目标的分类精度。
在具体实施时,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,步骤S105将语义分割模型和阈值分割模型进行模型融合,获取分类模型,具体可以包括:当语义分割模型的输出结果与阈值分割模型的输出结果均为水体或植被时,或,当语义分割模型的输出结果为裸土或不透水面且阈值分割模型的输出结果为其他时,判定分类结果正确;将判定为正确的分类结果作为新的地物分类样本集,使用迁移学习的方法继续训练语义分割模型,获取分类模型。这样使用语义分割对遥感影像进行分类,同时针对部分地物类型如水体、植被等,使用近红外波段特征进行提取,对两个结果进行决策级融合,然后使用半监督分类的方法增加样本,并再次分类,可以获取更高的分类精度。
具体地,步骤S105使用决策级融合的方法,语义分割模型的输出结果是一个二维矩阵,每个位置代表该像元的类别,总共包含四个类别,阈值分割的结果也是一个二维矩阵,每个位置代表该像元的类别,与语义分割不同的是,这里只包含三个类别,水体、植被、其它。对于每一个像素,当语义分割获取结果为裸土或不透水面,阈值分割获取为其他,或者都为水体或者植被,则认为分类正确。当分类结果不同时,对这部分不分类,将已经分类的作为新的训练样本,使用迁移学习的方法,继续训练语义分割模型,获取最终的分类结果。
在具体实施时,在本发明实施例提供的上述高分辨率遥感影像半监督分类方法中,步骤S106使用分类模型对待分类遥感影像进行分类,具体可以包括:首先,将待分类遥感影像切片,切片大小与步骤一相同,然后将每一个切片输入到分类模型中分类,经过膨胀、腐蚀等后处理操作,再把最后各个切片的分类结果拼接起来,添加坐标信息,生成tiff文件,实例如图7所示。
需要说明的是,本发明实施例提供的上述高分辨率遥感影像半监督分类方法主要是考虑高分辨率遥感影像的纹理特征和近红外波段的物理特征,通过改进的Unet++语义分割网络与阈值分割相结合的方式,采用半监督分类的方法,扩充训练样本,并再次训练语义分割模型以获取更高的分类精度,最后经过分类后处理,生成分类结果。
基于同一发明构思,本发明实施例还提供了一种高分辨率遥感影像半监督分类装置,由于该装置解决问题的原理与前述一种高分辨率遥感影像半监督分类方法相似,因此该装置的实施可以参见高分辨率遥感影像半监督分类方法的实施,重复之处不再赘述。
在具体实施时,本发明实施例提供的高分辨率遥感影像半监督分类装置,如图8所示,具体包括:
影像处理模块11,用于对遥感影像进行预处理;
样本集制作模块12,用于根据处理后的遥感影像,制作地物分类样本集;
第一模型构建模块13,用于构建基于Unet++网络的遥感影像语义分割模型,并通过地物分类样本集对语义分割模型进行训练;
第二模型构建模块14,用于构建基于近红外波段的阈值分割模型;
模型融合模块15,用于将语义分割模型和阈值分割模型进行模型融合,获取分类模型;
影像分类模块16,用于使用分类模型对待分类遥感影像进行分类。
在本发明实施例提供的上述高分辨率遥感影像半监督分类装置中,可以通过上述六个模块的相互作用,使遥感影像的纹理信息和近红外波段的光谱信息融合后,再对高分辨率遥感影像分类,提高了分类精度。
关于上述各个模块更加具体的工作过程可以参考前述实施例公开的相应内容,在此不再进行赘述。
相应地,本发明实施例还公开了一种高分辨率遥感影像半监督分类设备,包括处理器和存储器;其中,处理器执行存储器中存储的计算机程序 时实现前述实施例公开的高分辨率遥感影像半监督分类方法。
关于上述方法更加具体的过程可以参考前述实施例中公开的相应内容,在此不再进行赘述。
进一步地,本发明还公开了一种计算机可读存储介质,用于存储计算机程序;计算机程序被处理器执行时实现前述公开的高分辨率遥感影像半监督分类方法。
关于上述方法更加具体的过程可以参考前述实施例中公开的相应内容,在此不再进行赘述。
本说明书中各个实施例采用递进的方式描述,每个实施例重点说明的都是与其它实施例的不同之处,各个实施例之间相同或相似部分互相参见即可。对于实施例公开的装置、设备、存储介质而言,由于其与实施例公开的方法相对应,所以描述的比较简单,相关之处参见方法部分说明即可。
专业人员还可以进一步意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、计算机软件或者二者的结合来实现,为了清楚地说明硬件和软件的可互换性,在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
结合本文中所公开的实施例描述的方法或算法的步骤可以直接用硬件、处理器执行的软件模块,或者二者的结合来实施。软件模块可以置于随机存储器(RAM)、内存、只读存储器(ROM)、电可编程ROM、电可擦除可编程ROM、寄存器、硬盘、可移动磁盘、CD-ROM、或技术领域内所公知的任意其它形式的存储介质中。
综上,本发明实施例提供的一种高分辨率遥感影像半监督分类方法,包括:对遥感影像进行预处理;根据处理后的遥感影像,制作地物分类样本集;构建基于Unet++网络的遥感影像语义分割模型,并通过地物分类样本集对语义分割模型进行训练;构建基于近红外波段的阈值分割模型;将 语义分割模型和阈值分割模型进行模型融合,获取分类模型;使用分类模型对待分类遥感影像进行分类。这样通过构建基于Unet++网络的遥感影像语义分割模型和基于近红外波段的阈值分割模型,然后使用多模型融合的方法,使遥感影像的纹理信息和近红外波段的光谱信息融合,再对高分辨率遥感影像分类,能够提高分类精度。此外,本发明还针对高分辨率遥感影像半监督分类方法提供了相应的装置、设备及计算机可读存储介质,进一步使得上述方法更具有实用性,该装置、设备及计算机可读存储介质具有相应的优点。
最后,还需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
以上对本发明所提供的高分辨率遥感影像半监督分类方法、装置、设备及介质进行了详细介绍,本文中应用了具体个例对本发明的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本发明的方法及其核心思想;同时,对于本领域的一般技术人员,依据本发明的思想,在具体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本发明的限制。凡在本发明的精神和原理之内所作的任何修改、等同替换、改进等,均应包含在本发明的权利要求范围之内。

Claims (10)

  1. 一种高分辨率遥感影像半监督分类方法,其特征在于,包括:
    对遥感影像进行预处理;
    根据处理后的所述遥感影像,制作地物分类样本集;
    构建基于Unet++网络的遥感影像语义分割模型,并通过所述地物分类样本集对所述语义分割模型进行训练;
    构建基于近红外波段的阈值分割模型;
    将所述语义分割模型和所述阈值分割模型进行模型融合,获取分类模型;
    使用所述分类模型对待分类遥感影像进行分类。
  2. 根据权利要求1所述的高分辨率遥感影像半监督分类方法,其特征在于,所述对遥感影像进行预处理,包括:
    对遥感影像进行全色和多光谱影像融合、辐射校正、大气校正、几何校正的处理。
  3. 根据权利要求2所述的高分辨率遥感影像半监督分类方法,其特征在于,所述根据处理后的所述遥感影像,制作地物分类样本集,包括:
    建立与处理后的所述遥感影像相同大小的矢量,并将所述矢量全要素分割,分成水体、植被、裸土、不透水面四个类型;
    对分成的类型进行标注,并根据类型字段不同,转换为栅格数据;
    对处理后的所述遥感影像进行切片处理,同时裁剪出对应位置的标注的栅格,得到一组大小相同的图片和标签作为地物分类样本集。
  4. 根据权利要求3所述的高分辨率遥感影像半监督分类方法,其特征在于,所述Unet++网络采用多级上采样和跳层连接的方法,提取多层特征。
  5. 根据权利要求4所述的高分辨率遥感影像半监督分类方法,其特征在于,所述Unet++网络包括下采样层,上采样层,以及用于对所述下采样层进行特征抽取的中间层;其中,
    所述下采样层添加有EfficientB4模型的特征提取部分。
  6. 根据权利要求5所述的高分辨率遥感影像半监督分类方法,其特征在于,所述构建基于近红外波段的阈值分割模型,包括:
    使用阈值分割法获取处理后的所述遥感影像的近红外波段阈值直方图;
    将所述地物分类样本集和所述近红外波段阈值直方图进行匹配,选取出所述近红外波段阈值直方图中水体和植被的阈值区域,以构建阈值分割模型。
  7. 根据权利要求6所述的高分辨率遥感影像半监督分类方法,其特征在于,所述将所述语义分割模型和所述阈值分割模型进行模型融合,获取分类模型,包括:
    当所述语义分割模型的输出结果与所述阈值分割模型的输出结果均为水体或植被时,或,当所述语义分割模型的输出结果为裸土或不透水面且所述阈值分割模型的输出结果为其他时,判定分类结果正确;
    将判定为正确的分类结果作为新的地物分类样本集,使用迁移学习的方法继续训练所述语义分割模型,获取分类模型。
  8. 一种高分辨率遥感影像半监督分类装置,其特征在于,包括:
    影像处理模块,用于对遥感影像进行预处理;
    样本集制作模块,用于根据处理后的所述遥感影像,制作地物分类样本集;
    第一模型构建模块,用于构建基于Unet++网络的遥感影像语义分割模型,并通过所述地物分类样本集对所述语义分割模型进行训练;
    第二模型构建模块,用于构建基于近红外波段的阈值分割模型;
    模型融合模块,用于将所述语义分割模型和所述阈值分割模型进行模型融合,获取分类模型;
    影像分类模块,用于使用所述分类模型对待分类遥感影像进行分类。
  9. 一种高分辨率遥感影像半监督分类设备,其特征在于,包括处理器和存储器,其中,所述处理器执行所述存储器中存储的计算机程序时实现如权利要求1至7任一项所述的高分辨率遥感影像半监督分类方法。
  10. 一种计算机可读存储介质,其特征在于,用于存储计算机程序,其中,所述计算机程序被处理器执行时实现如权利要求1至7任一项所述的高分辨率遥感影像半监督分类方法。
PCT/CN2021/107289 2021-07-20 2021-07-20 高分辨率遥感影像半监督分类方法、装置、设备及介质 WO2023000159A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/107289 WO2023000159A1 (zh) 2021-07-20 2021-07-20 高分辨率遥感影像半监督分类方法、装置、设备及介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/107289 WO2023000159A1 (zh) 2021-07-20 2021-07-20 高分辨率遥感影像半监督分类方法、装置、设备及介质

Publications (1)

Publication Number Publication Date
WO2023000159A1 true WO2023000159A1 (zh) 2023-01-26

Family

ID=84979665

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/107289 WO2023000159A1 (zh) 2021-07-20 2021-07-20 高分辨率遥感影像半监督分类方法、装置、设备及介质

Country Status (1)

Country Link
WO (1) WO2023000159A1 (zh)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115965954A (zh) * 2023-03-16 2023-04-14 北京市农林科学院信息技术研究中心 秸秆类型识别方法、装置、电子设备及存储介质
CN115995005A (zh) * 2023-03-22 2023-04-21 航天宏图信息技术股份有限公司 基于单期高分辨率遥感影像的农作物的提取方法和装置
CN116129278A (zh) * 2023-04-10 2023-05-16 牧马人(山东)勘察测绘集团有限公司 一种基于遥感影像的土地利用分类识别***
CN116168301A (zh) * 2023-04-25 2023-05-26 耕宇牧星(北京)空间科技有限公司 一种基于嵌套编码器网络的农田施肥栅格检测方法
CN116503597A (zh) * 2023-04-26 2023-07-28 杭州芸起科技有限公司 一种跨域裸地语义分割网络构建方法、装置及存储介质
CN117095299A (zh) * 2023-10-18 2023-11-21 浙江省测绘科学技术研究院 破碎化耕作区的粮食作物提取方法、***、设备及介质
CN117110217A (zh) * 2023-10-23 2023-11-24 安徽农业大学 一种立体化水质监测方法及***
CN117349462A (zh) * 2023-12-06 2024-01-05 自然资源陕西省卫星应用技术中心 一种遥感智能解译样本数据集生成方法
CN117671519A (zh) * 2023-12-14 2024-03-08 上海勘测设计研究院有限公司 大区域遥感影像地物提取方法及***
CN117935079A (zh) * 2024-01-29 2024-04-26 珠江水利委员会珠江水利科学研究院 一种遥感影像融合方法、***及可读存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108960345A (zh) * 2018-08-08 2018-12-07 广东工业大学 一种遥感图像的融合方法、***及相关组件
WO2020240477A1 (en) * 2019-05-31 2020-12-03 Thales Canada Inc. Method and processing device for training a neural network
CN112560577A (zh) * 2020-11-13 2021-03-26 空间信息产业发展股份有限公司 一种基于语义分割的遥感图像地物分类方法
CN112613516A (zh) * 2020-12-11 2021-04-06 北京影谱科技股份有限公司 用于航拍视频数据的语义分割方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108960345A (zh) * 2018-08-08 2018-12-07 广东工业大学 一种遥感图像的融合方法、***及相关组件
WO2020240477A1 (en) * 2019-05-31 2020-12-03 Thales Canada Inc. Method and processing device for training a neural network
CN112560577A (zh) * 2020-11-13 2021-03-26 空间信息产业发展股份有限公司 一种基于语义分割的遥感图像地物分类方法
CN112613516A (zh) * 2020-12-11 2021-04-06 北京影谱科技股份有限公司 用于航拍视频数据的语义分割方法

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115965954A (zh) * 2023-03-16 2023-04-14 北京市农林科学院信息技术研究中心 秸秆类型识别方法、装置、电子设备及存储介质
CN115995005A (zh) * 2023-03-22 2023-04-21 航天宏图信息技术股份有限公司 基于单期高分辨率遥感影像的农作物的提取方法和装置
CN115995005B (zh) * 2023-03-22 2023-08-01 航天宏图信息技术股份有限公司 基于单期高分辨率遥感影像的农作物的提取方法和装置
CN116129278A (zh) * 2023-04-10 2023-05-16 牧马人(山东)勘察测绘集团有限公司 一种基于遥感影像的土地利用分类识别***
CN116168301A (zh) * 2023-04-25 2023-05-26 耕宇牧星(北京)空间科技有限公司 一种基于嵌套编码器网络的农田施肥栅格检测方法
CN116503597A (zh) * 2023-04-26 2023-07-28 杭州芸起科技有限公司 一种跨域裸地语义分割网络构建方法、装置及存储介质
CN117095299A (zh) * 2023-10-18 2023-11-21 浙江省测绘科学技术研究院 破碎化耕作区的粮食作物提取方法、***、设备及介质
CN117095299B (zh) * 2023-10-18 2024-01-26 浙江省测绘科学技术研究院 破碎化耕作区的粮食作物提取方法、***、设备及介质
CN117110217A (zh) * 2023-10-23 2023-11-24 安徽农业大学 一种立体化水质监测方法及***
CN117110217B (zh) * 2023-10-23 2024-01-12 安徽农业大学 一种立体化水质监测方法及***
CN117349462A (zh) * 2023-12-06 2024-01-05 自然资源陕西省卫星应用技术中心 一种遥感智能解译样本数据集生成方法
CN117349462B (zh) * 2023-12-06 2024-03-12 自然资源陕西省卫星应用技术中心 一种遥感智能解译样本数据集生成方法
CN117671519A (zh) * 2023-12-14 2024-03-08 上海勘测设计研究院有限公司 大区域遥感影像地物提取方法及***
CN117935079A (zh) * 2024-01-29 2024-04-26 珠江水利委员会珠江水利科学研究院 一种遥感影像融合方法、***及可读存储介质

Similar Documents

Publication Publication Date Title
WO2023000159A1 (zh) 高分辨率遥感影像半监督分类方法、装置、设备及介质
Song et al. Spatiotemporal satellite image fusion using deep convolutional neural networks
Zhu et al. Deep learning meets SAR: Concepts, models, pitfalls, and perspectives
CN110443143B (zh) 多分支卷积神经网络融合的遥感图像场景分类方法
Zhou et al. Pyramid fully convolutional network for hyperspectral and multispectral image fusion
Isikdogan et al. Seeing through the clouds with deepwatermap
CN111127374B (zh) 一种基于多尺度密集网络的Pan-sharpening方法
CN107358260B (zh) 一种基于表面波cnn的多光谱图像分类方法
Chang et al. Multisensor satellite image fusion and networking for all-weather environmental monitoring
CN112419155B (zh) 一种全极化合成孔径雷达影像超分辨率重建方法
CN113609889B (zh) 基于敏感特征聚焦感知的高分辨遥感影像植被提取方法
CN113312993B (zh) 一种基于PSPNet的遥感数据土地覆盖分类方法
CN115984714B (zh) 一种基于双分支网络模型的云检测方法
CN113420759B (zh) 一种基于深度学习的抗遮挡与多尺度死鱼识别***与方法
CN115511767B (zh) 一种自监督学习的多模态图像融合方法及其应用
CN112561876A (zh) 基于图像的池塘和水库的水质检测方法及***
CN112017192A (zh) 基于改进U-Net网络的腺体细胞图像分割方法及***
CN113516084B (zh) 高分辨率遥感影像半监督分类方法、装置、设备及介质
Dai et al. Gated convolutional networks for cloud removal from bi-temporal remote sensing images
CN113887472A (zh) 基于级联颜色及纹理特征注意力的遥感图像云检测方法
CN113591633A (zh) 基于动态自注意力Transformer的面向对象土地利用信息解译方法
CN113837941A (zh) 图像超分模型的训练方法、装置及计算机可读存储介质
He et al. Object‐Based Distinction between Building Shadow and Water in High‐Resolution Imagery Using Fuzzy‐Rule Classification and Artificial Bee Colony Optimization
CN117058367A (zh) 高分辨率遥感影像建筑物语义分割方法及装置
Rampi et al. Minnesota Land Cover Classification and Impervious Surface Area by Landsat and Lidar: 2013-14 Update

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21950425

Country of ref document: EP

Kind code of ref document: A1