Effects of Label Noise on Performance of Remote Sensing and Deep Learning-Based Water Body Segmentation Models

High spatial resolution remote sensing image (HSRRSI) data provide rich texture, geometric structure, and spatial distribution information for surface water bodies. The rich detail information provides better representation of the internal components of each object category and better reflects the relationships between adjacent objects. In this context, recognition methods such as geographic object-based image analysis (GEOBIA) have improved significantly. However, these methods focus mainly on bottom-up classifications from visual features to semantic categories, but ignore top-down feedback which can optimize recognition results. In recent years, deep learning has been applied in the field of remote sensing measurements because of its powerful feature extraction ability. A special convolutional neural network (CNN) based region proposal generation and object detection integrated framework has greatly improved the performance of object detection for HSRRSI, which provides a new method for water body recognition based on remote sensing data. This study uses the excellent “self-learning ability” of deep learning to construct a modified structure of the Mask R-CNN method which integrates bottom-up and top-down processes for water recognition. Compared with traditional methods, our method is completely data-driven without prior knowledge, and it can be regarded as a novel technical procedure for water body recognition in practical engineering application. Experimental results indicate that the method produces accurate recognition results for multi-source and multi-temporal water bodies, and can effectively avoid confusion with shadows and other ground features.

Download Full-text

Deep Learning Network Intensification for Preventing Noisy-Labeled Samples for Remote Sensing Classification

Remote Sensing ◽

10.3390/rs13091689 ◽

2021 ◽

Vol 13 (9) ◽

pp. 1689

Author(s):

Chuang Lin ◽

Shanxin Guo ◽

Jinsong Chen ◽

Luyi Sun ◽

Xiaorou Zheng ◽

...

Keyword(s):

Remote Sensing ◽

Deep Learning ◽

Network Performance ◽

Aerial Image ◽

Label Noise ◽

Remote Sensing Classification ◽

Learning Network ◽

Training Samples ◽

Deep Learning Network ◽

The Impact

The deep-learning-network performance depends on the accuracy of the training samples. The training samples are commonly labeled by human visual investigation or inherited from historical land-cover or land-use maps, which usually contain label noise, depending on subjective knowledge and the time of the historical map. Helping the network to distinguish noisy labels during the training process is a prerequisite for applying the model for training across time and locations. This study proposes an antinoise framework, the Weight Loss Network (WLN), to achieve this goal. The WLN contains three main parts: (1) the segmentation subnetwork, which any state-of-the-art segmentation network can replace; (2) the attention subnetwork (λ); and (3) the class-balance coefficient (α). Four types of label noise (an insufficient label, redundant label, missing label and incorrect label) were simulated by dilate and erode processing to test the network’s antinoise ability. The segmentation task was set to extract buildings from the Inria Aerial Image Labeling Dataset, which includes Austin, Chicago, Kitsap County, Western Tyrol and Vienna. The network’s performance was evaluated by comparing it with the original U-Net model by adding noisy training samples with different noise rates and noise levels. The result shows that the proposed antinoise framework (WLN) can maintain high accuracy, while the accuracy of the U-Net model dropped. Specifically, after adding 50% of dilated-label samples at noise level 3, the U-Net model’s accuracy dropped by 12.7% for OA, 20.7% for the Mean Intersection over Union (MIOU) and 13.8% for Kappa scores. By contrast, the accuracy of the WLN dropped by 0.2% for OA, 0.3% for the MIOU and 0.8% for Kappa scores. For eroded-label samples at the same level, the accuracy of the U-Net model dropped by 8.4% for OA, 24.2% for the MIOU and 43.3% for Kappa scores, while the accuracy of the WLN dropped by 4.5% for OA, 4.7% for the MIOU and 0.5% for Kappa scores. This result shows that the antinoise framework proposed in this paper can help current segmentation models to avoid the impact of noisy training labels and has the potential to be trained by a larger remote sensing image set regardless of the inner label error.

Download Full-text

Water-Body Segmentation for Multi-Spectral Remote Sensing Images by Feature Pyramid Enhancement and Pixel Pair Matching

International Journal of Remote Sensing ◽

10.1080/01431161.2021.1906981 ◽

2021 ◽

Vol 42 (13) ◽

pp. 5029-5047

Author(s):

Suting Chen ◽

Yao Liu ◽

Chuang Zhang

Keyword(s):

Remote Sensing ◽

Water Body ◽

Remote Sensing Images ◽

Pixel Pair ◽

Body Segmentation ◽

Feature Pyramid ◽

Pair Matching

Download Full-text

Semantic Segmentation of Satellite Images using Deep Learning

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.h9186.0610821 ◽

2021 ◽

Vol 10 (8) ◽

pp. 33-37

Author(s):

Chandra Pal Kushwah ◽

Kuruna Markam

Keyword(s):

Remote Sensing ◽

Deep Learning ◽

Deep Neural Networks ◽

Semantic Segmentation ◽

Remote Sensing Image ◽

Learning Performance ◽

Test Results ◽

Natural Scene ◽

Target Segmentation ◽

Segmentation Models

Bidirectional in recent years, Deep learning performance in natural scene image processing has improved its use in remote sensing image analysis. In this paper, we used the semantic segmentation of remote sensing images for deep neural networks (DNN). To make it ideal for multi-target semantic segmentation of remote sensing image systems, we boost the Seg Net encoder-decoder CNN structures with index pooling & U-net. The findings reveal that the segmentation of various objects has its benefits and drawbacks for both models. Furthermore, we provide an integrated algorithm that incorporates two models. The test results indicate that the integrated algorithm proposed will take advantage of all multi-target segmentation models and obtain improved segmentation relative to two models.

Download Full-text

Research and Application of Automatic Recognition and Extraction of Water Body in Remote Sensing Images Based on Deep Learning

10.1007/978-981-16-5168-7_22 ◽

2021 ◽

pp. 167-173

Author(s):

Bo Wu ◽

Yi Gao ◽

Xiaolei Xie ◽

Ye Yang

Keyword(s):

Remote Sensing ◽

Deep Learning ◽

Water Body ◽

Automatic Recognition ◽

Remote Sensing Images

Download Full-text

A Deep Learning Method of Water Body Extraction From High Resolution Remote Sensing Images With Multisensors

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing ◽

10.1109/jstars.2021.3060769 ◽

2021 ◽

Vol 14 ◽

pp. 3120-3132

Author(s):

Mengya Li ◽

Penghai Wu ◽

Biao Wang ◽

Honglyun Park ◽

Yang Hui ◽

...

Keyword(s):

Remote Sensing ◽

Deep Learning ◽

High Resolution ◽

Water Body ◽

Learning Method ◽

Remote Sensing Images

Download Full-text

A Deep Learning-Based Framework for Automated Extraction of Building Footprint Polygons from Very High-Resolution Aerial Imagery

Remote Sensing ◽

10.3390/rs13183630 ◽

2021 ◽

Vol 13 (18) ◽

pp. 3630

Author(s):

Ziming Li ◽

Qinchuan Xin ◽

Ying Sun ◽

Mengying Cao

Keyword(s):

Remote Sensing ◽

Deep Learning ◽

High Resolution ◽

Semantic Segmentation ◽

Aerial Imagery ◽

Learning Models ◽

Building Footprint ◽

Bounding Boxes ◽

Very High ◽

Segmentation Models

Accurate building footprint polygons provide essential data for a wide range of urban applications. While deep learning models have been proposed to extract pixel-based building areas from remote sensing imagery, the direct vectorization of pixel-based building maps often leads to building footprint polygons with irregular shapes that are inconsistent with real building boundaries, making it difficult to use them in geospatial analysis. In this study, we proposed a novel deep learning-based framework for automated extraction of building footprint polygons (DLEBFP) from very high-resolution aerial imagery by combining deep learning models for different tasks. Our approach uses the U-Net, Cascade R-CNN, and Cascade CNN deep learning models to obtain building segmentation maps, building bounding boxes, and building corners, respectively, from very high-resolution remote sensing images. We used Delaunay triangulation to construct building footprint polygons based on the detected building corners with the constraints of building bounding boxes and building segmentation maps. Experiments on the Wuhan University building dataset and ISPRS Vaihingen dataset indicate that DLEBFP can perform well in extracting high-quality building footprint polygons. Compared with the other semantic segmentation models and the vector map generalization method, DLEBFP is able to achieve comparable mapping accuracies with semantic segmentation models on a pixel basis and generate building footprint polygons with concise edges and vertices with regular shapes that are close to the reference data. The promising performance indicates that our method has the potential to extract accurate building footprint polygons from remote sensing images for applications in geospatial analysis.

Download Full-text

Deep Learning Based High-Resolution Remote Sensing Image classification

International Journal of Advanced Research in Computer Science and Software Engineering ◽

10.23956/ijarcsse.v7i10.384 ◽

2017 ◽

Vol 7 (10) ◽

pp. 22

Author(s):

Sumit Kaur

Keyword(s):

Machine Learning ◽

Remote Sensing ◽

Deep Learning ◽

Image Classification ◽

Language Processing ◽

Object Perception ◽

Remote Sensing Image ◽

Research Area ◽

Remote Sensing Image Classification ◽

Unsupervised Algorithms

Abstract- Deep learning is an emerging research area in machine learning and pattern recognition field which has been presented with the goal of drawing Machine Learning nearer to one of its unique objectives, Artificial Intelligence. It tries to mimic the human brain, which is capable of processing and learning from the complex input data and solving different kinds of complicated tasks well. Deep learning (DL) basically based on a set of supervised and unsupervised algorithms that attempt to model higher level abstractions in data and make it self-learning for hierarchical representation for classification. In the recent years, it has attracted much attention due to its state-of-the-art performance in diverse areas like object perception, speech recognition, computer vision, collaborative filtering and natural language processing. This paper will present a survey on different deep learning techniques for remote sensing image classification.

Download Full-text