A review on early wildfire detection from unmanned aerial vehicles using deep learning-based computer vision algorithms

Automatic extraction of the road surface and road centerline from very high-resolution (VHR) remote sensing images has always been a challenging task in the field of feature extraction. Most existing road datasets are based on data with simple and clear backgrounds under ideal conditions, such as images derived from Google Earth. Therefore, the studies on road surface extraction and road centerline extraction under complex scenes are insufficient. Meanwhile, most existing efforts addressed these two tasks separately, without considering the possible joint extraction of road surface and centerline. With the introduction of multitask convolutional neural network models, it is possible to carry out these two tasks simultaneously by facilitating information sharing within a multitask deep learning model. In this study, we first design a challenging dataset using remote sensing images from the GF-2 satellite. The dataset contains complex road scenes with manually annotated images. We then propose a two-task and end-to-end convolution neural network, termed Multitask Road-related Extraction Network (MRENet), for road surface extraction and road centerline extraction. We take features extracted from the road as the condition of centerline extraction, and the information transmission and parameter sharing between the two tasks compensate for the potential problem of insufficient road centerline samples. In the network design, we use atrous convolutions and a pyramid scene parsing pooling module (PSP pooling), aiming to expand the network receptive field, integrate multilevel features, and obtain more abundant information. In addition, we use a weighted binary cross-entropy function to alleviate the background imbalance problem. Experimental results show that the proposed algorithm outperforms several comparative methods in the aspects of classification precision and visual interpretation.

https://www.mdpi.com/2072-4292/13/2/239/pdf

MRENet: Simultaneous Extraction of Road Surface and Road Centerline in Complex Urban Scenes from Very High-Resolution Images

Deep learning and convolutional neural network technologies are increasingly used in the problems of analysis, segmentation and recognition of objects in images. In this article a convolutional neural network for automated wildfire detection on high-resolution aerial photos is presented. Two databases of satellite RGB-images with different spatial resolution containing 1457 and 393 high-resolution images, respectively, were prepared for training and testing the neural network. Various techniques of data augmentation are used to enlarge training and test sets generated by data windowing. U-Net neural network with the ResNet34 as encoder was used in research. Neural network training was learning using the NVIDIA DGX-1 supercomputer. Adaptive moment estimation algorithm was used for optimization of training process. Special metrics, such as Sorensen-Dice coefficient, precision, recall, F1-score and IoU value allows to measure the quality of developed model. The developed algorithm can be successfully applied for early wildland fires detection in practical applications.

Wildfire Segmentation on Satellite Images using Deep Learning

Cell phones have become a necessity for many people throughout the world. The ability to keep in touch with family, business associates, and access to email are only a few of the reasons for the increasing importance of cell phones. However, the mobile-phones in early times were bulky, restrictive to only some features and worked only in areas where there was a good connection. All these problems were resolved by integrating a processor within a cell-phone. The processor is the central hub of your smartphone. It receives and executes every command, performing billions of calculations per second. The effectiveness of the processor directly affects every application you run, whether it's the camera, the music player, or just a simple email program. In the following journal, we have discussed the distinctive features, merits, and demerits of the latest mobile phone processors of different Tech companies.

A Comparative Study on the Recent Smart Mobile Phone Processors

Early stage smoke detection using image and video analysis is an important area of research due to its enormous applications in mitigating fire hazards and ensuring environmental safety. Numerous solutions have been proposed for real-time smoke detection using conventional image processing, machine learning, and deep learning techniques. Smoke pattern, motion analysis, color and texture are important characteristics that help identify it in the outdoor environment. Vision-based Smoke detection algorithms can be broadly classified into three categories: smoke classification, segmentation, and bounding box estimation. This paper presents a comprehensive survey of existing techniques on smoke detection in the outdoor environment using image and video analysis. To perform the survey, initially 271 articles were collected from different sources like Google Scholar, Science Direct, IEEE Xplore, SpringerLink, Wiley and ACM Digital Library using the keyword search. Based on their focus on the vision-based solutions for the outdoor environment, 126 articles were identified as relevant to the present survey. Starting from the initial IP approaches that are frequently referred in the literature, machine learning and deep learning approaches have also been reviewed for each type of smoke detection. Performance of algorithms, datasets used in the research, evaluation metrics, challenges and future directions of research are also discussed. 

A survey on vision-based outdoor smoke detection techniques for environmental safety

Results of training of convolutional neural network for satellite four-channel image segmentation are performed. Input images contain blue, green, red and near-infrared channels. The algorithm was trained to detect buildings and other urban areas. Modification of the U-Net neural network with two encoders was used. The values of Sorensen coefficient and Jaccard index were calculated for 16 different urban regions.

Modification of U-Net neural network in the task of multichannel satellite images segmentation

The aim of this research is to create a deep learning algorithm for automated forest areas segmentation on high-resolution aerial images. Loss values and Dice coefficient, which compares results of algorithms with real masks, was used to measure the quality of developed model. On the whole this paper demonstrates how convolutional neural network implemented on modern GPUs can be applied for the detection of forests on satellite images.

Forest Areas Segmentation on Aerial Images by Deep Learning

This article presents results of two deep learning algorithms for sand quarries detection on high-resolution aerial photos. Planet database of high-resolution aerial images was collected. Input images contain blue, green, red and near-infrared channels. Before the training process there was implemented the equalization of brightness histogram which allows to cope with the problem of obscuration of satellite images. To implement numerical experiments there were extracted smaller patches. Training and test sets were enlarged by three types of data augmentation. Convolutional neural networks were pretrained on the SpaceNet dataset and tuned on the Planet database. Deep learning algorithms were launched on NVIDIA DGX-1 supercomputer. Special metrics, such as F1, precision, recall and Dice coefficient allowed to compare the quality of developed models.

Quarry Areas Segmentation on Satellite Images by Convolutional Neural Networks

Deep learning and modern type of neural network technologies are increasingly used for the detection, segmentation and classification of different objects in aerial multichannel images. The goal of given research was to develop a deep learning algorithm for automated building detection on four-channel satellite images. It is proposed to use U-Net neural network with two decoders to separate objects, one decoder is trained to segment buildings and structures, and the other detects narrow distances between buildings. It is shown that optimized U-Net can be used to detect such kind of objects efficiently. The model was implemented by means of open Keras library and launched on modern GPUs of high-performance supercomputer NVIDIA DGX-1. Before testing of developed algorithm on Planet aerial image dataset, modified U-Net had been pre-trained on SpaceNet database. The process of image data augmentation is described. The problem of effective building detection on high-resolution aerial photos can be used for urban planning, building control, search of the best locations for future outlets etc.

Roman Larionov

Papers

Wildfire Segmentation on Satellite Images using Deep Learning

Modification of U-Net neural network in the task of multichannel satellite images segmentation

Forest Areas Segmentation on Aerial Images by Deep Learning

Quarry Areas Segmentation on Satellite Images by Convolutional Neural Networks

Separation of Closely Located Buildings on Aerial Images Using U-Net Neural Network