Analyzing satellite images by apply deep learning instance segmentation of agricultural fields (original) (raw)
Related papers
Edge Detection for Satellite Images without Deep Networks
2021
Satellite imagery is widely used in many application sectors, including agriculture, navigation, and urban planning. Frequently, satellite imagery involves both large numbers of images as well as high pixel counts, making satellite datasets computationally expensive to analyze. Recent approaches to satellite image analysis have largely emphasized deep learning methods. Though extremely powerful, deep learning has some drawbacks, including the requirement of specialized computing hardware and a high reliance on training data. When dealing with large satellite datasets, the cost of both computational resources and training data annotation may be prohibitive. In this paper, we demonstrate that a carefully designed image pre-processing pipeline allows traditional computer vision techniques to achieve semantic edge detection in satellite imagery that is competitive with deep learning methods at a fraction of the resource costs. We focus on the task of semantic edge detection due to its g...
Insights on Deep Learning based Segmentation Schemes Towards Analyzing Satellite Imageries
Satellite imageries are essentially a complex form of an image when subjected to critical analytical operation. The analytical process applied on remotely sensed satellite imageries are utilized for generating the land cover map. With an abundance of traditional techniques evolved to date, deep learning-based schemes are progressively gaining pace for identifying and classifying a terrestrial object in satellite images. However, different variants of deep learning approaches have different operations, and so are the consequences. At the same time, there is no reported literature to highlight the issues, trends, and effectiveness much on a generalized scale concerning segmentation. Therefore, this paper reviews some of the recent segmentation approaches using deep learning to contribute towards review findings in the form of research trends, research gaps, and essential learning outcomes. The paper offers a compact and distinct picture of deep learning approaches used to boost segmentation for satellite images.
Farm Area Segmentation in Satellite Images Using DeepLabv3+ Neural Networks
2019
Farm detection using low resolution satellite images is an important part of digital agriculture applications such as crop yield monitoring. However, it has not received enough attention compared to high-resolution images. Although high resolution images are more efficient for detection of land cover components, the analysis of low-resolution images are yet important due to the low-resolution repositories of the past satellite images used for timeseries analysis, free availability and economic concerns. In this paper, semantic segmentation of farm areas is addressed using low resolution satellite images. The segmentation is performed in two stages; First, local patches or Regions of Interest (ROI) that include farm areas are detected. Next, deep semantic segmentation strategies are employed to detect the farm pixels. For patch classification, two previously developed local patch classification strategies are employed; a two-step semi-supervised methodology using hand-crafted feature...
Deep Learning for Understanding Satellite Imagery: An Experimental Survey
Frontiers in Artificial Intelligence, 2020
Translating satellite imagery into maps requires intensive effort and time, especially leading to inaccurate maps of the affected regions during disaster and conflict. The combination of availability of recent datasets and advances in computer vision made through deep learning paved the way toward automated satellite image translation. To facilitate research in this direction, we introduce the Satellite Imagery Competition using a modified SpaceNet dataset. Participants had to come up with different segmentation models to detect positions of buildings on satellite images. In this work, we present five approaches based on improvements of U-Net and Mask R-Convolutional Neuronal Networks models, coupled with unique training adaptations using boosting algorithms, morphological filter, Conditional Random Fields and custom losses. The good results—as high as AP=0.937 and AR=0.959—from these models demonstrate the feasibility of Deep Learning in automated satellite image annotation.
Satellite-Net: Automatic Extraction of Land Cover Indicators from Satellite Imagery by Deep Learning
ArXiv, 2019
In this paper we address the challenge of land cover classification for satellite images via Deep Learning (DL). Land Cover aims to detect the physical characteristics of the territory and estimate the percentage of land occupied by a certain category of entities: vegetation, residential buildings, industrial areas, forest areas, rivers, lakes, etc. DL is a new paradigm for Big Data analytics and in particular for Computer Vision. The application of DL in images classification for land cover purposes has a great potential owing to the high degree of automation and computing performance. In particular, the invention of Convolution Neural Networks (CNNs) was a fundament for the advancements in this field. In [1], the Satellite Task Team of the UN Global Working Group describes the results achieved so far with respect to the use of earth observation for Official Statistics. However, in that study, CNNs have not yet been explored for automatic classification of imagery. This work invest...
Comparative Analysis on Deep Learning based Pan-sharpening of Very High-Resolution Satellite Images
International journal of environment and geoinformatics, 2021
Pan-sharpening is a fundamental task of remote sensing, aiming to produce a synthetic image having high spatial and spectral resolution of original panchromatic and multispectral images. In recent years, as in other tasks of the remote sensing field, deep learning based approaches have been developed for this task. In this research, a detailed comparative analysis was conducted to evaluate the performance and visual quality of pan-sharpening results from traditional algorithms and deep learning-based models. For this purpose, the deep learning based methods that are CNN based pan-sharpening (PNN), Multiscale and multi-depth convolutional neural networks (MSDCNN) and Pan-sharpened Generative Adversarial Networks (PSGAN) and traditional methods that are Brovey, PCA, HIS, Indusion and PRACS were applied. Analysis was performed on regions with different land cover characteristics to evaluate the stability of the methods. In addition, effects of the filter size, spectral indices, activation and loss functions on the pan-sharpening were investigated. For the accuracy assessment, commonly used with-reference and withoutreference quality metrics were computed in addition to visual quality evaluations. According to results, the deep learning-based methods provided promising results in both the reduced resolution and full resolution experiments, while PRACS method outperformed other traditional algorithms in most of the experimental configurations.
In 2021 The 4th International Conference on Machine Learning and Machine Intelligence (MLMI’21), September 17–19, 2021, Hangzhou, China. ACM, New York, NY, USA, 8 pages., 2021
Deep learning classification has state-of-the-art machine learning approaches. Earlier work proves the deep convolutional neural network was successful and brilliant in different tasks such as image classification and image processing in remote sensing datasets. To recognize and clarify the physical aspect of the earth's surface land cover and exploit the land use is an exciting issue in environmental monitoring analyzing remote sensing data that is free and easy to get in different areas without time-consuming. To improve the quality of data sources and cooperating with image representation is still lack methods. We started with the traditional segmentation approach to divide an image into a single and compare multi-level into multiple segmentation for remote sensing imagery. We review the traditional machine learning and deep learning neural network classification tasks. We collected a dataset from six countries of Eastern Africa Communities with nine categories. We proposed an ensemble average model neural network with three models, combined as a single model trained, and achieved 96.5% validation accuracy. Finally, we compared state-of-the-art and extracted features pre-trained weight ImageNet and used traditional machine learning algorithms to improve accuracy.
International Journal of Advanced Research in Computer Science, 2017
Geographical satellite images that are used for the analysis of environmental and geographical plains are obtained through remote sensing techniques. The raw images collected from the satellites are not well suited for statistical analysis and accurate report preparation. So, the raw images undergo the usual image processing procedure such as preprocessing, segmentation, feature extraction and classification. Traditional image classification techniques have several spatial and spectral resolution issues. A novel image classification technique, namely, Convolutional Neural Networks (CNN) technique is an emerging research criterion. It is an extension of neural networks and deep learning approaches. In this paper, several CNN based image classification techniques are analyzed and their performance is compared. The techniques involved in this analysis include Full Convolutional Network (FCN), Patch-based classification, pixel-to-pixel based segmentation and convnet-based feature extrac...
Feature Extraction from Satellite Images Using Segnet and Fully Convolutional Networks (FCN)
International Journal of Engineering and Geosciences, 2020
Object detection and classification are among the most popular topics in Photogrammetry and Remote Sensing studies. With technological developments, a large number of high-resolution satellite images have been obtained and it has become possible to distinguish many different objects. Despite all these developments, the need for human intervention in object detection and classification is seen as one of the major problems. Machine learning has been used as a priority option to this day to reduce this need. Although success has been achieved with this method, human intervention is still needed. Deep learning provides a great convenience by eliminating this problem. Deep learning methods carry out the learning process on raw data unlike traditional machine learning methods. Although deep learning has a long history, the main reasons for its increased popularity in recent years are; the availability of sufficient data for the training proce ss and the availability of hardware to process the data. In this study, a performance comparison was made between two different convolutional neural network architectures (SegNet and Fully Convolutional Networks (FCN)) which are used for object segmentation and classification on images. These two different models were trained using the same training dataset and their performances have been evaluated using the same test dataset. The results show that, for building segmentation, there is not much significant difference between these two architectures in terms of accuracy, but FCN architecture is more successful than SegNet by 1%. However, this situation may vary according to the dataset used during the training of the system.