scholarly journals Improving the Model for Person Detection in Aerial Image Sequences Using the Displacement Vector: A Search and Rescue Scenario

Drones ◽  
2022 ◽  
Vol 6 (1) ◽  
pp. 19
Author(s):  
Mirela Kundid Vasić ◽  
Vladan Papić

Recent results in person detection using deep learning methods applied to aerial images gathered by Unmanned Aerial Vehicles (UAVs) have demonstrated the applicability of this approach in scenarios such as Search and Rescue (SAR) operations. In this paper, the continuation of our previous research is presented. The main goal is to further improve detection results, especially in terms of reducing the number of false positive detections and consequently increasing the precision value. We present a new approach that, as input to the multimodel neural network architecture, uses sequences of consecutive images instead of only one static image. Since successive images overlap, the same object of interest needs to be detected in more than one image. The correlation between successive images was calculated, and detected regions in one image were translated to other images based on the displacement vector. The assumption is that an object detected in more than one image has a higher probability of being a true positive detection because it is unlikely that the detection model will find the same false positive detections in multiple images. Based on this information, three different algorithms for rejecting detections and adding detections from one image to other images in the sequence are proposed. All of them achieved precision value about 80% which is increased by almost 20% compared to the current state-of-the-art methods.

Author(s):  
S. Su ◽  
T. Nawata ◽  
T. Fuse

Abstract. Automatic building change detection has become a topical issue owing to its wide range of applications, such as updating building maps. However, accurate building change detection remains challenging, particularly in urban areas. Thus far, there has been limited research on the use of the outdated building map (the building map before the update, referred to herein as the old-map) to increase the accuracy of building change detection. This paper presents a novel deep-learning-based method for building change detection using bitemporal aerial images containing RGB bands, bitemporal digital surface models (DSMs), and an old-map. The aerial images have two types of spatial resolutions, 12.5 cm or 16 cm, and the cell size of the DSMs is 50 cm × 50 cm. The bitemporal aerial images, the height variations calculated using the differences between the bitemporal DSMs, and the old-map were fed into a network architecture to build an automatic building change detection model. The performance of the model was quantitatively and qualitatively evaluated for an urban area that covered approximately 10 km2 and contained over 21,000 buildings. The results indicate that it can detect the building changes with optimum accuracy as compared to other methods that use inputs such as i) bitemporal aerial images only, ii) bitemporal aerial images and bitemporal DSMs, and iii) bitemporal aerial images and an old-map. The proposed method achieved recall rates of 89.3%, 88.8%, and 99.5% for new, demolished, and other buildings, respectively. The results also demonstrate that the old-map is an effective data source for increasing building change detection accuracy.


Electronics ◽  
2020 ◽  
Vol 9 (9) ◽  
pp. 1459
Author(s):  
Mirela Kundid Vasić ◽  
Vladan Papić

In this paper, we propose a novel method for person detection in aerial images of nonurban terrain gathered by an Unmanned Aerial Vehicle (UAV), which plays an important role in Search And Rescue (SAR) missions. The UAV in SAR operations contributes significantly due to the ability to survey a larger geographical area from an aerial viewpoint. Because of the high altitude of recording, the object of interest (person) covers a small part of an image (around 0.1%), which makes this task quite challenging. To address this problem, a multimodel deep learning approach is proposed. The solution consists of two different convolutional neural networks in region proposal, as well as in the classification stage. Additionally, contextual information is used in the classification stage in order to improve the detection results. Experimental results tested on the HERIDAL dataset achieved precision of 68.89% and a recall of 94.65%, which is better than current state-of-the-art methods used for person detection in similar scenarios. Consequently, it may be concluded that this approach is suitable for usage as an auxiliary method in real SAR operations.


Aerial images provide a landscape view of earth surfaces that utilized to monitor the large areas. Each Aerial image comprises the different scenes to identify the objects on the digital maps. The several methodologies have been developed to solve the problem of the scene classification using input aerial images. The method does not improve the classification performance using more aerial images. In order to improve the classification performance, a Tanimoto Gaussian Kernelized Feature Extraction Based Multinomial GentleBoost Classification (TGKFE-MGBC) technique is introduced. The TGKFE-MGBC technique comprises three major processes namely object-based segmentation, feature extraction and aerial image scene classification. At first, object-based segmentation partitions the aerial image into several sub-bands. Aerial image with more than two objects is called as multi-spectral. The objects in spectral bands are identified by Tanimoto pixel similarity measure. This process helps to reduce the feature extraction time. Each object has different features like shape, size, color, texture and so on. After that, Gaussian Kernelized Feature Extraction is carried out to extracts the features from the objects with minimal time. Finally, the Multinomial GentleBoost Classification is applied for categorizing the scenes into different classes with the extracted features. The GentleBoost is an ensemble technique uses multinomial naïve Bayes probabilistic classifier as a weak learner and it combines to makes a strong one for classifying the scenes. The strong classifier result improves the aerial image scene classification accuracy and minimizes the false positive rate. Simulation is conducted using aerial image database with different factors such as feature extraction time, aerial image scene classification accuracy and false positive rate. The results showed that the TGKFE-MGBC technique effectively improves the aerial image scene classification accuracy and minimizes the feature extraction time as well as the false positive rate.


2012 ◽  
Vol 225 ◽  
pp. 310-314 ◽  
Author(s):  
Mohamad Mahmud Zihad ◽  
Kamarul Arifin Ahmad ◽  
A. Halim Kadarman

This paper presents an ongoing study and research of a 2-Axis stabilized aerial image capturing system to obtain aerial images. Aerial images are commonly used for reconnaissance, area surveying, and also for search and rescue mission. Currently, several methods of remote sensing were developed with multiple objectives either for civil or military applications to obtain high precision images. The study involves the design and fabrication of 2-Axis stabilized image system platform. Rolling and pitching motion of an air vehicle effects while airborne to acquire sharp vertical images are the main consideration in this study.


2020 ◽  
Vol 2020 (10) ◽  
pp. 54-62
Author(s):  
Oleksii VASYLIEV ◽  

The problem of applying neural networks to calculate ratings used in banking in the decision-making process on granting or not granting loans to borrowers is considered. The task is to determine the rating function of the borrower based on a set of statistical data on the effectiveness of loans provided by the bank. When constructing a regression model to calculate the rating function, it is necessary to know its general form. If so, the task is to calculate the parameters that are included in the expression for the rating function. In contrast to this approach, in the case of using neural networks, there is no need to specify the general form for the rating function. Instead, certain neural network architecture is chosen and parameters are calculated for it on the basis of statistical data. Importantly, the same neural network architecture can be used to process different sets of statistical data. The disadvantages of using neural networks include the need to calculate a large number of parameters. There is also no universal algorithm that would determine the optimal neural network architecture. As an example of the use of neural networks to determine the borrower's rating, a model system is considered, in which the borrower's rating is determined by a known non-analytical rating function. A neural network with two inner layers, which contain, respectively, three and two neurons and have a sigmoid activation function, is used for modeling. It is shown that the use of the neural network allows restoring the borrower's rating function with quite acceptable accuracy.


2017 ◽  
Vol 14 (4) ◽  
pp. 70-78
Author(s):  
M. А. Epifanov

The article presents brief results of the analysis of the subscriber segment of the international space search and rescue system COSPAS-SARSAT. Proposals on the nomenclature of emergency subscriber emergency terminals and direction finders are presented, which are necessary for development and production in batch production in Russia, in order to prevent a backlog in this area and to implement the import substitution program. The recommendations are developed taking into account the materials discussed at the working technical groups, the Joint Committee and the COSPAS-SARSAT Council in recent years, as well as the results of the technical analysis of the nomenclature of terminals produced and developed by foreign companies. In addition, the development of recommendations takes into account the main current and prospective directions for the development of Russia's economy and industry.


2020 ◽  
Vol 22 (5) ◽  
pp. 51-55
Author(s):  
OLEG N. KORCHAGIN ◽  
◽  
ANASTASIA V. LYADSKAYA ◽  

The article is devoted to the current state of digitalization aimed at solving urgent problems of combating corruption in the field of public administration and private business sector. The work considers the experience of foreign countries and the influence of digital technologies on the fight against corruption. It is noted that the digitalization of public administration is becoming one of the decisive factors for increasing the efficiency of the anti-corruption system and improving management mechanisms. Big Data, if integrated and structured according to the given parameters, allows the implementation of legislative, law enforcement, control and supervisory and law enforcement activities reliably and transparently. Big Data tools allow us to analyze processes, identify dependencies and predict corruption risks. The author describes the most significant problems that complicate the transfer of offline technologies into the online environment. The paper analyzes promising directions for the development of digital technologies that would lead to solving the arising problems, as well as to implement tasks that previously seemed unreachable. The article also describes current developments in the field of collecting and managing large amounts of data, the “Internet of Things”, modern network architecture, and other advances in the field of IT; the work provides applied examples of their potential use in the field of combating corruption. The study gives reasons that, in the context of combating corruption, digitalization should be allocated in a separate area of activity that is controlled and regulated by the state.


2019 ◽  
Vol 11 (10) ◽  
pp. 1157 ◽  
Author(s):  
Jorge Fuentes-Pacheco ◽  
Juan Torres-Olivares ◽  
Edgar Roman-Rangel ◽  
Salvador Cervantes ◽  
Porfirio Juarez-Lopez ◽  
...  

Crop segmentation is an important task in Precision Agriculture, where the use of aerial robots with an on-board camera has contributed to the development of new solution alternatives. We address the problem of fig plant segmentation in top-view RGB (Red-Green-Blue) images of a crop grown under open-field difficult circumstances of complex lighting conditions and non-ideal crop maintenance practices defined by local farmers. We present a Convolutional Neural Network (CNN) with an encoder-decoder architecture that classifies each pixel as crop or non-crop using only raw colour images as input. Our approach achieves a mean accuracy of 93.85% despite the complexity of the background and a highly variable visual appearance of the leaves. We make available our CNN code to the research community, as well as the aerial image data set and a hand-made ground truth segmentation with pixel precision to facilitate the comparison among different algorithms.


Sign in / Sign up

Export Citation Format

Share Document