Data fusion techniques have gained special interest in remote sensing due to the available capabilities to obtain measurements from the same scene using different instruments with varied resolution domains. In particular, multispectral (MS) and hyperspectral (HS) imaging fusion is used to generate high spatial and spectral images (HSEI). Deep learning data fusion models based on Long Short Term Memory (LSTM) and Convolutional Neural Networks (CNN) have been developed to achieve such task.In this work, we present a Multi-Level Propagation Learning Network (MLPLN) based on a LSTM model but that can be trained with variable data sizes in order achieve the fusion process. Moreover, the MLPLN provides an intrinsic data augmentation feature that reduces the required number of training samples. The proposed model generates a HSEI by fusing a high-spatial resolution MS image and a low spatial resolution HS image. The performance of the model is studied and compared to existing CNN and LSTM approaches by evaluating the quality of the fused image using the structural similarity metric (SSIM). The results show that an increase in the SSIM is still obtained while reducing of the number of training samples to train the MLPLN model.
more »
« less
Analyzing the effects of pixel-scale data fusion in hyperspectral image classification performance
Recently, multispectral and hyperspectral data fusion models based on deep learning have been proposed to generate images with a high spatial and spectral resolution. The general objective is to obtain images that improve spatial resolution while preserving high spectral content. In this work, two deep learning data fusion techniques are characterized in terms of classification accuracy. These methods fuse a high spatial resolution multispectral image with a lower spatial resolution hyperspectral image to generate a high spatial-spectral hyperspectral image. The first model is based on a multi-scale long short-term memory (LSTM) network. The LSTM approach performs the fusion using a multiple step process that transitions from low to high spatial resolution using an intermediate step capable of reducing spatial information loss while preserving spectral content. The second fusion model is based on a convolutional neural network (CNN) data fusion approach. We present fused images using four multi-source datasets with different spatial and spectral resolutions. Both models provide fused images with increased spatial resolution from 8m to 1m. The obtained fused images using the two models are evaluated in terms of classification accuracy on several classifiers: Minimum Distance, Support Vector Machines, Class-Dependent Sparse Representation and CNN classification. The classification results show better performance in both overall and average accuracy for the images generated with the multi-scale LSTM fusion over the CNN fusion
more »
« less
- Award ID(s):
- 1750970
- NSF-PAR ID:
- 10224973
- Editor(s):
- Messinger, David W.; Velez-Reyes, Miguel
- Date Published:
- Journal Name:
- SPIE 11392, Algorithms, Technologies, and Applications for Multispectral and Hyperspectral Imagery XXVI
- Page Range / eLocation ID:
- 3
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Pixel-level fusion of satellite images coming from multiple sensors allows for an improvement in the quality of the acquired data both spatially and spectrally. In particular, multispectral and hyperspectral images have been fused to generate images with a high spatial and spectral resolution. In literature, there are several approaches for this task, nonetheless, those techniques still present a loss of relevant spatial information during the fusion process. This work presents a multi scale deep learning model to fuse multispectral and hyperspectral data, each with high-spatial-and-low-spectral resolution (HSaLS) and low-spatial-and-high-spectral resolution (LSaHS) respectively. As a result of the fusion scheme, a high-spatial-and-spectral resolution image (HSaHS) can be obtained. In order of accomplishing this result, we have developed a new scalable high spatial resolution process in which the model learns how to transition from low spatial resolution to an intermediate spatial resolution level and finally to the high spatial-spectral resolution image. This step-by-step process reduces significantly the loss of spatial information. The results of our approach show better performance in terms of both the structural similarity index and the signal to noise ratio.more » « less
-
Messinger, David W. ; Velez-Reyes, Miguel (Ed.)Recent advances in data fusion provide the capability to obtain enhanced hyperspectral data with high spatial and spectral information content, thus allowing for an improved classification accuracy. Although hyperspectral image classification is a highly investigated topic in remote sensing, each classification technique presents different advantages and disadvantages. For example; methods based on morphological filtering are particularly good at classifying human-made structures with basic geometrical spatial shape, like houses and buildings. On the other hand, methods based on spectral information tend to perform better classification in natural scenery with more shape diversity such as vegetation and soil areas. Even more, for those classes with mixed pixels, small training data or objects with similar re ectance values present a higher challenge to obtain high classification accuracy. Therefore, it is difficult to find just one technique that provides the highest accuracy of classification for every class present in an image. This work proposes a decision fusion approach aiming to increase classification accuracy of enhanced hyperspectral images by integrating the results of multiple classifiers. Our approach is performed in two-steps: 1) the use of machine learning algorithms such as Support Vector Machines (SVM), Deep Neural Networks (DNN) and Class-dependent Sparse Representation will generate initial classification data, then 2) the decision fusion scheme based on a Convolutional Neural Network (CNN) will integrate all the classification results into a unified classification rule. In particular, the CNN receives as input the different probabilities of pixel values from each implemented classifier, and using a softmax activation function, the final decision is estimated. We present results showing the performance of our method using different hyperspectral image datasets.more » « less
-
The spatial distribution of forest stands is one of the fundamental properties of forests. Timely and accurately obtained stand distribution can help people better understand, manage, and utilize forests. The development of remote sensing technology has made it possible to map the distribution of tree species in a timely and accurate manner. At present, a large amount of remote sensing data have been accumulated, including high-spatial-resolution images, time-series images, light detection and ranging (LiDAR) data, etc. However, these data have not been fully utilized. To accurately identify the tree species of forest stands, various and complementary data need to be synthesized for classification. A curve matching based method called the fusion of spectral image and point data (FSP) algorithm was developed to fuse high-spatial-resolution images, time-series images, and LiDAR data for forest stand classification. In this method, the multispectral Sentinel-2 image and high-spatial-resolution aerial images were first fused. Then, the fused images were segmented to derive forest stands, which are the basic unit for classification. To extract features from forest stands, the gray histogram of each band was extracted from the aerial images. The average reflectance in each stand was calculated and stacked for the time-series images. The profile curve of forest structure was generated from the LiDAR data. Finally, the features of forest stands were compared with training samples using curve matching methods to derive the tree species. The developed method was tested in a forest farm to classify 11 tree species. The average accuracy of the FSP method for ten performances was between 0.900 and 0.913, and the maximum accuracy was 0.945. The experiments demonstrate that the FSP method is more accurate and stable than traditional machine learning classification methods.more » « less
-
State-of-the-art deep learning technology has been successfully applied to relatively small selected areas of very high spatial resolution (0.15 and 0.25 m) optical aerial imagery acquired by a fixed-wing aircraft to automatically characterize ice-wedge polygons (IWPs) in the Arctic tundra. However, any mapping of IWPs at regional to continental scales requires images acquired on different sensor platforms (particularly satellite) and a refined understanding of the performance stability of the method across sensor platforms through reliable evaluation assessments. In this study, we examined the transferability of a deep learning Mask Region-Based Convolutional Neural Network (R-CNN) model for mapping IWPs in satellite remote sensing imagery (~0.5 m) covering 272 km2 and unmanned aerial vehicle (UAV) (0.02 m) imagery covering 0.32 km2. Multi-spectral images were obtained from the WorldView-2 satellite sensor and pan-sharpened to ~0.5 m, and a 20 mp CMOS sensor camera onboard a UAV, respectively. The training dataset included 25,489 and 6022 manually delineated IWPs from satellite and fixed-wing aircraft aerial imagery near the Arctic Coastal Plain, northern Alaska. Quantitative assessments showed that individual IWPs were correctly detected at up to 72% and 70%, and delineated at up to 73% and 68% F1 score accuracy levels for satellite and UAV images, respectively. Expert-based qualitative assessments showed that IWPs were correctly detected at good (40–60%) and excellent (80–100%) accuracy levels for satellite and UAV images, respectively, and delineated at excellent (80–100%) level for both images. We found that (1) regardless of spatial resolution and spectral bands, the deep learning Mask R-CNN model effectively mapped IWPs in both remote sensing satellite and UAV images; (2) the model achieved a better accuracy in detection with finer image resolution, such as UAV imagery, yet a better accuracy in delineation with coarser image resolution, such as satellite imagery; (3) increasing the number of training data with different resolutions between the training and actual application imagery does not necessarily result in better performance of the Mask R-CNN in IWPs mapping; (4) and overall, the model underestimates the total number of IWPs particularly in terms of disjoint/incomplete IWPs.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1117/12.2560106
