scholarly journals Drone to Obstacle Distance Estimation Using YOLO V3 Network and Mathematical Principles

2022 ◽  
Vol 2161 (1) ◽  
pp. 012022
Author(s):  
N Aswini ◽  
S V Uma ◽  
V Akhilesh
Keyword(s):  
Distance Estimation ◽  
Traffic Monitoring ◽  
Obstacle Detection ◽  
Construction Sites ◽  
Positioning Systems ◽  
Pipeline Inspection ◽  
Detection Model ◽  
Hazardous Environments ◽  
Learning Techniques ◽  
Source Of Information

Abstract Now a days, drones are very commonly used in various real time applications. Moving towards autonomy, these drones rely on obstacle detection sensors and various collision avoidance algorithms programmed into it. Development of fully autonomous drones provide the fundamental benefits of being able to operate in hazardous environments without a human pilot. Among the various sensors, monocular cameras provide a rich source of information and are one of the main sensing mechanisms in low flying drones. These drones can be used for rescue and search operations, traffic monitoring, infrastructure, and pipeline inspection, and in construction sites. In this paper, we propose an onboard obstacle detection model using deep learning techniques, combined with a mathematical approach to calculate the distance between the detected obstacle and the drone. This when implemented does not need any additional sensor or Global Positioning Systems (GPS) other than the vision sensor.

scholarly journals High-Precision RTT-Based Indoor Positioning System Using RCDN and RPN

Sensors ◽  
2021 ◽  
Vol 21 (11) ◽  
pp. 3701
Author(s):  
Ju-Hyeon Seong ◽  
Soo-Hwan Lee ◽  
Won-Yeol Kim ◽  
Dong-Hoan Seo
Keyword(s):  
High Precision ◽  
Distance Estimation ◽  
Sampling Period ◽  
Multipath Fading ◽  
Indoor Positioning ◽  
Positioning System ◽  
Positioning Systems ◽  
Positioning Errors ◽  
Location Determination ◽  
Indoor Positioning System

Wi-Fi round-trip timing (RTT) was applied to indoor positioning systems based on distance estimation. RTT has a higher reception instability than the received signal strength indicator (RSSI)-based fingerprint in non-line-of-sight (NLOS) environments with many obstacles, resulting in large positioning errors due to multipath fading. To solve these problems, in this paper, we propose high-precision RTT-based indoor positioning system using an RTT compensation distance network (RCDN) and a region proposal network (RPN). The proposed method consists of a CNN-based RCDN for improving the prediction accuracy and learning rate of the received distances and a recurrent neural network-based RPN for real-time positioning, implemented in an end-to-end manner. The proposed RCDN collects and corrects a stable and reliable distance prediction value from each RTT transmitter by applying a scanning step to increase the reception rate of the TOF-based RTT with unstable reception. In addition, the user location is derived using the fingerprint-based location determination method through the RPN in which division processing is applied to the distances of the RTT corrected in the RCDN using the characteristics of the fast-sampling period.

scholarly journals A Neural Network Approach for Building An Obstacle Detection Model by Fusion of Proximity Sensors Data

Sensors ◽  
2018 ◽  
Vol 18 (3) ◽  
pp. 683 ◽  
Author(s):  
Gonzalo Farias ◽  
Ernesto Fabregas ◽  
Emmanuel Peralta ◽  
Héctor Vargas ◽  
Gabriel Hermosilla ◽  
...  
Keyword(s):  
Neural Network ◽  
Obstacle Detection ◽  
Network Approach ◽  
Neural Network Approach ◽  
Detection Model

scholarly journals Acquiring Sentiment from Twitter using Supervised Learning and Lexicon-based Techniques

2016 ◽  
Vol 15 (1) ◽  
pp. 63-80
Author(s):  
Jitrlada ROJRATANAVIJIT ◽  
Preecha VICHITTHAMAROS ◽  
Sukanya PHONGSUPHAP
Keyword(s):  
Supervised Learning ◽  
Mixed Method ◽  
Short Length ◽  
Writing Systems ◽  
Learning Techniques ◽  
Average Accuracy ◽  
The Difference ◽  
Tweet Classification ◽  
Source Of Information ◽  
Processing Steps

The emergence of Twitter in Thailand has given millions of users a platform to express and share their opinions about products and services, among other subjects, and so Twitter is considered to be a rich source of information for companies to understand their customers by extracting and analyzing sentiment from Tweets. This offers companies a fast and effective way to monitor public opinions on their brands, products, services, etc. However, sentiment analysis performed on Thai Tweets has challenges brought about by language-related issues, such as the difference in writing systems between Thai and English, short-length messages, slang words, and word usage variation. This research paper focuses on Tweet classification and on solving data sparsity issues. We propose a mixed method of supervised learning techniques and lexicon-based techniques to filter Thai opinions and to then classify them into positive, negative, or neutral sentiments. The proposed method includes a number of pre-processing steps before the text is fed to the classifier. Experimental results showed that the proposed method overcame previous limitations from other studies and was very effective in most cases. The average accuracy was 84.80 %, with 82.42 % precision, 83.88 % recall, and 82.97 % F-measure.

scholarly journals Sole Mate: Safe Path-Finding by Obstacle Detection and Distance Estimation for the Blind

2019 ◽  
Vol 8 (6S4) ◽  
pp. 50-54
Keyword(s):  
Haptic Feedback ◽  
Distance Estimation ◽  
Feedback System ◽  
Obstacle Detection ◽  
Video Footage ◽  
System A ◽  
The World ◽  
Distance Sensor ◽  
Visually Challenged ◽  
Additional Device

The world has increased its demand for assistive technology (AT). There are a lot of researches and developments going on with respect to AT. Among the AT devices which are being developed, the need for a reliable and less expensive device which serves as an assistance for a visually challenged person is in serious demand all around the world. We, therefore, intend to provide a solution for this by constructing a device that has the capability to detect the obstacles within a given range for a visually challenged person and alerting the person about the obstacles. This involves various components like a camera for image detection, an ultrasonic distance sensor for distance estimation and a vibration motor which works on the principle of Haptic feedback and rotates with varied intensities depending on how far the obstacle is from the user. This paper presents a model which is a part of the footwear of the user and hence, no additional device is required to hold onto for assistance. The model involves the use of a microcontroller, a camera, to dynamically perceive the obstacles and a haptic feedback system to alert the person about the same. The camera dynamically acquires the real time video footage which is further processed by the microcontroller to detect the obstacles. Simultaneously, one more algorithm is being executed to estimate the distance with the help of an ultrasonic distance sensor. Depending on the distance, the frequency of the vibration motor, which acts as the output for notifying the user about the obstacle, is varied (haptic feedback). With this system, a visually challenged person will be able to avoid the obstacles successfully without the use of any additional device.

Machine Learning Techniques for IoT-Based Indoor Tracking and Localization

2022 ◽  
pp. 123-145
Author(s):  
Pelin Yildirim Taser ◽  
Vahid Khalilpour Akram
Keyword(s):  
Machine Learning ◽  
Indoor Localization ◽  
Learning Algorithms ◽  
Distance Estimation ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
The Internet ◽  
Traditional Methods ◽  
Learning Techniques ◽  
Localization Systems

The GPS signals are not available inside the buildings; hence, indoor localization systems rely on indoor technologies such as Bluetooth, WiFi, and RFID. These signals are used for estimating the distance between a target and available reference points. By combining the estimated distances, the location of the target nodes is determined. The wide spreading of the internet and the exponential increase in small hardware diversity allow the creation of the internet of things (IoT)-based indoor localization systems. This chapter reviews the traditional and machine learning-based methods for IoT-based positioning systems. The traditional methods include various distance estimation and localization approaches; however, these approaches have some limitations. Because of the high prediction performance, machine learning algorithms are used for indoor localization problems in recent years. The chapter focuses on presenting an overview of the application of machine learning algorithms in indoor localization problems where the traditional methods remain incapable.

scholarly journals Ship Detection from X-Band SAR Images Using M2Det Deep Learning Model

2020 ◽  
Vol 10 (21) ◽  
pp. 7751
Author(s):  
Seong-Jae Hong ◽  
Won-Kyung Baek ◽  
Hyung-Sup Jung
Keyword(s):  
Deep Learning ◽  
Learning Process ◽  
Intensity Difference ◽  
Sar Images ◽  
Ship Detection ◽  
Detection Model ◽  
Learning Techniques ◽  
Vessel Detection ◽  
Intensity Images ◽  
Deep Learning Model

Synthetic aperture radar (SAR) images have been used in many studies for ship detection because they can be captured without being affected by time and weather. In recent years, the development of deep learning techniques has facilitated studies on ship detection in SAR images using deep learning techniques. However, because the noise from SAR images can negatively affect the learning of the deep learning model, it is necessary to reduce the noise through preprocessing. In this study, deep learning vessel detection was performed using preprocessed SAR images, and the effects of the preprocessing of the images on deep learning vessel detection were compared and analyzed. Through the preprocessing of SAR images, (1) intensity images, (2) decibel images, and (3) intensity difference and texture images were generated. The M2Det object detection model was used for the deep learning process and preprocessed SAR images. After the object detection model was trained, ship detection was performed using test images. The test results are presented in terms of precision, recall, and average precision (AP), which were 93.18%, 91.11%, and 89.78% for the intensity images, respectively, 94.16%, 94.16%, and 92.34% for the decibel images, respectively, and 97.40%, 94.94%, and 95.55% for the intensity difference and texture images, respectively. From the results, it can be found that the preprocessing of the SAR images can facilitate the deep learning process and improve the ship detection performance. The results of this study are expected to contribute to the development of deep learning-based ship detection techniques in SAR images in the future.

scholarly journals Intact Detection of Highly Occluded Immature Tomatoes on Plants Using Deep Learning Techniques

Sensors ◽  
2020 ◽  
Vol 20 (10) ◽  
pp. 2984
Author(s):  
Yue Mu ◽  
Tai-Shen Chen ◽  
Seishi Ninomiya ◽  
Wei Guo
Keyword(s):  
Deep Learning ◽  
Low Cost ◽  
Yield Prediction ◽  
Learning Approaches ◽  
Size Estimation ◽  
Detection Model ◽  
Lighting Conditions ◽  
Learning Techniques ◽  
Location Map ◽  
Tomato Detection

Automatic detection of intact tomatoes on plants is highly expected for low-cost and optimal management in tomato farming. Mature tomato detection has been wildly studied, while immature tomato detection, especially when occluded with leaves, is difficult to perform using traditional image analysis, which is more important for long-term yield prediction. Therefore, tomato detection that can generalize well in real tomato cultivation scenes and is robust to issues such as fruit occlusion and variable lighting conditions is highly desired. In this study, we build a tomato detection model to automatically detect intact green tomatoes regardless of occlusions or fruit growth stage using deep learning approaches. The tomato detection model used faster region-based convolutional neural network (R-CNN) with Resnet-101 and transfer learned from the Common Objects in Context (COCO) dataset. The detection on test dataset achieved high average precision of 87.83% (intersection over union ≥ 0.5) and showed a high accuracy of tomato counting (R2 = 0.87). In addition, all the detected boxes were merged into one image to compile the tomato location map and estimate their size along one row in the greenhouse. By tomato detection, counting, location and size estimation, this method shows great potential for ripeness and yield prediction.

An improved multipurpose field robot for installing construction materials

Robotica ◽  
2010 ◽  
Vol 28 (7) ◽  
pp. 945-957 ◽  
Author(s):  
Seungyeol Lee ◽  
Seungnam Yu ◽  
Seokjong Yu ◽  
Changsoo Han
Keyword(s):  
Force Feedback ◽  
Construction Materials ◽  
Human Robot Interaction ◽  
Work Environments ◽  
National Defense ◽  
Construction Sites ◽  
Robot Interaction ◽  
Hazardous Environments ◽  
Remote Sites ◽  
Field Robot

SUMMARYRecently, there has been a lot of interest concerning remote-controlled robot manipulation in hazardous environments including construction sites, national defense areas, and disaster areas. However, there are problems involving the method of remote control in unstructured work environments such as construction sites. In a previous study, to address these problems, a multipurpose field robot (MFR) system was described. Though the case studies on construction, to which “MFR for installing construction materials” was applied, however, we found some factors to be improved. In this paper, we introduce a prototype of improved multipurpose field robot (IMFR) for construction work. This prototype robot helps a human operator easily install construction materials in remote sites through an upgraded additional module. This module consists of a force feedback joystick and a monitoring device. The human–robot interaction and bilateral communication for strategic control is also described. To evaluate the proposed IMFR, the installation of construction materials was simulated. We simulated the process of installing construction materials, in this case a glass panel. The IMFR was expected to do more accurate work, safely, at construction sites as well as at environmentally hazardous areas that are difficult for humans to approach.

Classifying the Influential Individuals in Multi-Layer Social Networks

2019 ◽  
Vol 8 (1) ◽  
pp. 21-32
Author(s):  
Ruchi Mittal ◽  
M.P.S Bhatia
Keyword(s):  
Social Networks ◽  
Social Network ◽  
Information Diffusion ◽  
Real Life ◽  
Single Layer ◽  
Machine Learning Techniques ◽  
Centrality Measures ◽  
Learning Techniques ◽  
Modes Of Interaction ◽  
Source Of Information

Nowadays, social media is one of the popular modes of interaction and information diffusion. It is commonly found that the main source of information diffusion is done by some entities and such entities are also called as influencers. An influencer is an entity or individual who has the ability to influence others because of his/her relationship or connection with his/her audience. In this article, we propose a methodology to classify influencers from multi-layer social networks. A multi-layer social network is the same as a single layer social network depict that it includes multiple properties of a node and modeled them into multiple layers. The proposed methodology is a fusion of machine learning techniques (SVM, neural networks and so on) with centrality measures. We demonstrate the proposed algorithm on some real-life networks to validate the effectiveness of the approach in multi-layer systems.

Sign in / Sign up

Export Citation Format

Share Document