Cloud detection is an inextricable pre-processing step in remote sensing image analysis workflows. Most of the traditional rule-based and machine-learning-based algorithms utilize low-level features of the clouds and classify individual cloud pixels based on their spectral signatures. Cloud detection using such approaches can be challenging due to a multitude of factors including harsh lighting conditions, the presence of thin clouds, the context of surrounding pixels, and complex spatial patterns. In recent studies, deep convolutional neural networks (CNNs) have shown outstanding results in the computer vision domain. These methods are practiced for better capturing the texture, shape as well as context of images. In this study, we propose a deep learning CNN approach to detect cloud pixels from medium-resolution satellite imagery. The proposed CNN accounts for both the low-level features, such as color and texture information as well as high-level features extracted from successive convolutions of the input image. We prepared a cloud-pixel dataset of approximately 7273 randomly sampled 320 by 320 pixels image patches taken from a total of 121 Landsat-8 (30m) and Sentinel-2 (20m) image scenes. These satellite images come with cloud masks. From the available data channels, only blue, green, red, and NIR bands are fed into the model. The CNN model was trained on 5300 image patches and validated on 1973 independent image patches. As the final output from our model, we extract a binary mask of cloud pixels and non-cloud pixels. The results are benchmarked against established cloud detection methods using standard accuracy metrics.
more »
« less
Insights and approaches using deep learning to classify wildlife
The implementation of intelligent software to identify and classify objects and individuals in visual fields is a technology of growing importance to operatives in many fields, including wildlife conservation and management. To non-experts, the methods can be abstruse and the results mystifying. Here, in the context of applying cutting edge methods to classify wildlife species from camera-trap data, we shed light on the methods themselves and types of features these methods extract to make efficient identifications and reliable classifications. The current state of the art is to employ convolutional neural networks (CNN) encoded within deep-learning algorithms. We outline these methods and present results obtained in training a CNN to classify 20 African wildlife species with an overall accuracy of 87.5% from a dataset containing 111,467 images. We demonstrate the application of a gradient-weighted class-activation-mapping (Grad-CAM) procedure to extract the most salient pixels in the final convolution layer. We show that these pixels highlight features in particular images that in some cases are similar to those used to train humans to identify these species. Further, we used mutual information methods to identify the neurons in the final convolution layer that consistently respond most strongly across a set of images of one particular species. We then interpret the features in the image where the strongest responses occur, and present dataset biases that were revealed by these extracted features. We also used hierarchical clustering of feature vectors (i.e., the state of the final fully-connected layer in the CNN) associated with each image to produce a visual similarity dendrogram of identified species. Finally, we evaluated the relative unfamiliarity of images that were not part of the training set when these images were one of the 20 species “known” to our CNN in contrast to images of the species that were “unknown” to our CNN.
more » « less- NSF-PAR ID:
- 10154203
- Publisher / Repository:
- Nature Publishing Group
- Date Published:
- Journal Name:
- Scientific Reports
- Volume:
- 9
- Issue:
- 1
- ISSN:
- 2045-2322
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
In the past few years, there have been many research studies conducted in the field of Satellite Image Classification. The purposes of these studies included flood identification, forest fire monitoring, greenery land identification, and land-usage identification. In this field, finding suitable data is often considered problematic, and some research has also been done to identify and extract suitable datasets for classification. Although satellite data can be challenging to deal with, Convolutional Neural Networks (CNNs), which consist of multiple interconnected neurons, have shown promising results when applied to satellite imagery data. In the present work, first we have manually downloaded satellite images of four different classes in Florida locations using the TerraFly Mapping System, developed and managed by the High Performance Database Research Center at Florida International University. We then develop a CNN architecture suitable for extracting features and capable of multi-class classification in our dataset. We discuss the shortcomings in the classification due to the limited size of the dataset. To address this issue, we first employ data augmentation and then utilize transfer learning methodology for feature extraction with VGG16 and ResNet50 pretrained models. We use these features to classify satellite imagery of Florida. We analyze the misclassification in our model and, to address this issue, we introduce a location-based CNN model. We convert coordinates to geohash codes, use these codes as an additional feature vector and feed them into the CNN model. We believe that the new CNN model combined with geohash codes as location features provides a better accuracy for our dataset.
-
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
-
Classification and Feature Extraction for Hydraulic Structures Data Using Advanced CNN ArchitecturesAn efficient feature selection method can significantly boost results in classification problems. Despite ongoing improvement, hand-designed methods often fail to extract features capturing high- and mid-level representations at effective levels. In machine learning (Deep Learning), recent developments have improved upon these hand-designed methods by utilizing automatic extraction of features. Specifically, Convolutional Neural Networks (CNNs) are a highly successful technique for image classification which can automatically extract features, with ongoing learning and classification of these features. The purpose of this study is to detect hydraulic structures (i.e., bridges and culverts) that are important to overland flow modeling and environmental applications. The dataset used in this work is a relatively small dataset derived from 1-m LiDAR-derived Digital Elevation Models (DEMs) and National Agriculture Imagery Program (NAIP) aerial imagery. The classes for our experiment consist of two groups: the ones with a bridge/culvert being present are considered "True", and those without a bridge/culvert are considered "False". In this paper, we use advanced CNN techniques, including Siamese Neural Networks (SNNs), Capsule Networks (CapsNets), and Graph Convolutional Networks (GCNs), to classify samples with similar topographic and spectral characteristics, an objective which is challenging utilizing traditional machine learning techniques, such as Support Vector Machine (SVM), Gaussian Classifier (GC), and Gaussian Mixture Model (GMM). The advanced CNN-based approaches combined with data pre-processing techniques (e.g., data augmenting) produced superior results. These approaches provide efficient, cost-effective, and innovative solutions to the identification of hydraulic structures.more » « less
-
Implementing local contextual guidance principles in a single-layer CNN architecture, we propose an efficient algorithm for developing broad-purpose representations (i.e., representations transferable to new tasks without additional training) in shallow CNNs trained on limited-size datasets. A contextually guided CNN (CG-CNN) is trained on groups of neighboring image patches picked at random image locations in the dataset. Such neighboring patches are likely to have a common context and therefore are treated for the purposes of training as belonging to the same class. Across multiple iterations of such training on different context-sharing groups of image patches, CNN features that are optimized in one iteration are then transferred to the next iteration for further optimization, etc. In this process, CNN features acquire higher pluripotency, or inferential utility for any arbitrary classification task. In our applications to natural images and hyperspectral images, we find that CG-CNN can learn transferable features similar to those learned by the first layers of the well-known deep networks and produce favorable classification accuracies.more » « less
