Realization of real-time detection algorithms for key parts of unmanned aerial vehicle based on support vector machine

2020 ◽  
Author(s):  
Weixiang Wang ◽  
Hanyu Hong ◽  
Yaozong Zhang ◽  
Shiyang Li ◽  
Jiaowei Shi
Keyword(s):  
Support Vector Machine ◽  
Real Time ◽  
Unmanned Aerial Vehicle ◽  
Support Vector ◽  
Detection Algorithms ◽  
Aerial Vehicle ◽  
Real Time Detection

scholarly journals Real-Time Detection of Ground Objects Based on Unmanned Aerial Vehicle Remote Sensing with Deep Learning: Application in Excavator Detection for Pipeline Safety

2020 ◽  
Vol 12 (1) ◽  
pp. 182 ◽  
Author(s):  
Lingxuan Meng ◽  
Zhixing Peng ◽  
Ji Zhou ◽  
Jirong Zhang ◽  
Zhenyu Lu ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Deep Learning ◽  
Object Detection ◽  
Real Time ◽  
Unmanned Aerial Vehicle ◽  
Detection System ◽  
Detection Model ◽  
Aerial Vehicle ◽  
Real Time Detection ◽  
Pipeline Safety

Unmanned aerial vehicle (UAV) remote sensing and deep learning provide a practical approach to object detection. However, most of the current approaches for processing UAV remote-sensing data cannot carry out object detection in real time for emergencies, such as firefighting. This study proposes a new approach for integrating UAV remote sensing and deep learning for the real-time detection of ground objects. Excavators, which usually threaten pipeline safety, are selected as the target object. A widely used deep-learning algorithm, namely You Only Look Once V3, is first used to train the excavator detection model on a workstation and then deployed on an embedded board that is carried by a UAV. The recall rate of the trained excavator detection model is 99.4%, demonstrating that the trained model has a very high accuracy. Then, the UAV for an excavator detection system (UAV-ED) is further constructed for operational application. UAV-ED is composed of a UAV Control Module, a UAV Module, and a Warning Module. A UAV experiment with different scenarios was conducted to evaluate the performance of the UAV-ED. The whole process from the UAV observation of an excavator to the Warning Module (350 km away from the testing area) receiving the detection results only lasted about 1.15 s. Thus, the UAV-ED system has good performance and would benefit the management of pipeline safety.

scholarly journals Support Vector Machine Algorithm for Real-Time Detection of VF Signals

2011 ◽  
Vol 8 ◽  
pp. 602-608
Author(s):  
Chunyun Zhang ◽  
Jie Zhao ◽  
Fei Li ◽  
Huilin Jia ◽  
Jie Tian
Keyword(s):  
Support Vector Machine ◽  
Real Time ◽  
Support Vector ◽  
Support Vector Machine Algorithm ◽  
Real Time Detection

scholarly journals KLASIFIKASI MANGROVE BERBASIS OBJEK DAN PIKSEL MENGGUNAKAN CITRA SENTINEL-2B DI SUNGAI LIONG, BENGKALIS, PROVINSI RIAU

2018 ◽  
Vol 10 (3) ◽  
pp. 601-615
Author(s):  
. Rosmasita ◽  
Vincentius P. Siregar ◽  
Syamsul B. Agus
Keyword(s):  
Support Vector Machine ◽  
Land Cover ◽  
Unmanned Aerial Vehicle ◽  
Spatial Data ◽  
Data Availability ◽  
Support Vector ◽  
Object Based ◽  
Svm Algorithm ◽  
Aerial Vehicle ◽  
Increase In Accuracy

ABSTRAK Penelitian pemetaan mangrove di Sungai Liong, Bengkalis Provinsi Riau sangat terbatas, sehingga ketersediaan data spasial di wilayah ini masih sangat terbatas. Pemanfaatan citra satelit dapat dijadikan alternatif dalam menyediakan data spasial secara efektif dan efesien. Penelitian ini bertujuan untuk memetakan mangrove sampai tingkat komunitas menggunakan citra sentinel 2B dengan metode klasifikasi berbasis objek/OBIA dan membandingkannya dengan teknik klasifikasi berbasis piksel. Algoritma yang digunakan pada penelitian ini adalah support vector machine (SVM). Pengembangan skema klasifikasi mangrove pada penelitian ini di bagi menjadi 2 level, yaitu kelas penutup lahan di sekitar mangrove dan kelas komunitas mangrove. Data yang digunakan untuk klasifikasi kelas penutup lahan adalah data foto udara yang diperoleh dengan menggunakan pesawat tanpa awak (unmanned aerial vehicle/UAV) dan untuk klasifikasi komunitas menggunakan data transek tahun 2013. Akurasi keseluruhan  (OA) yang diperoleh untuk klafikasi penutup lahan mangrove dengan kedua teknik klasifikasi berbasis objek dan piksel berturut-turut adalah 78,7% dan 70,9%. Sedangkan akurasi keseluruhan (OA) untuk klasifikasi komunitas mangrove berbasis objek dan piksel berutru-turut yaitu 76,6% dan 75,0%. Sekitar 7,8% peningkatan akurasi pemetaan penutup lahan dan sekitar 1,6% peningkatan akurasi pemetaan komunitas mangrove yang diperoleh dengan metode klasifikasi berbasis objek. ABSTRACTResearch on mangrove mapping at the Liong River Bengkalis Riau Province was very limited, therefore the spatial data availability of mangrove in Liong River is also very limited. The use of satellite remote sensing to map mangrove has become widespread as it can provide accurate, effecient, and repeatable assessments. The purposed of this study was to map mangrove at the community level using sentinel 2B imagery based on object-based classification method (OBIA) and it compared pixel-based classification at Liong River, Bengkalis, Riau Provinc. This study was used support vector machine (SVM) algorithm. The scheme classification use is that land cover and mangrove community. The classification data of land cover was collected using unmanned aerial vehicle (UAV) and community mangrove was using transect data of 2013. The result of land cover classification and community mangrove indicated that object-based classification technique was better than pixel-based classification. The highest an overall accuracy of land cover is 78.7% versus 70.9%, whereas mangrove community is 76.6 versus 75.0%. Approximately 7.8% increase in accuracy can be achieved by object-based method of classification for land cover and 1.6% for mangrove community.

Design of a new low-cost unmanned aerial vehicle and vision-based concrete crack inspection method

2020 ◽  
Vol 19 (6) ◽  
pp. 1871-1883 ◽  
Author(s):  
Bin Lei ◽  
Yali Ren ◽  
Ning Wang ◽  
Linsheng Huo ◽  
Gangbing Song
Keyword(s):  
Real Time ◽  
Unmanned Aerial Vehicle ◽  
Concrete Structure ◽  
Low Cost ◽  
Central Point ◽  
Point Method ◽  
Great Promise ◽  
Support Vector ◽  
Inspection Method ◽  
Aerial Vehicle

With the explosive development of the computer vision technology, more and more vision-based inspection methods enabled by unmanned aerial vehicle technologies have been researched on the crack inspection of the sundry concrete structures. However, because of the limitation of the low-cost unmanned aerial vehicle hardware, whose cost is around US$500, most of the vision-based methods are difficult to be implemented on the low-cost unmanned aerial vehicle for real-time crack inspection. To address this challenge, in this article, a new computationally efficient vision-based crack inspection method is designed and successfully implemented on a low-cost unmanned aerial vehicle. Furthermore, to reduce the acquired data samples, a new algorithm entitled crack central point method is designed to extract the effective information from the pre-processed images. The proposed vision-based crack detection method includes the following three major components: (1) the image pre-processing algorithm, (2) crack central point method, and (3) the support vector machine model–based classifier. To demonstrate the effectiveness of the new inspection method, a concrete structure inspection experiment is implemented. The experimental results indicate that this new method is able to accurately and rapidly inspect the cracks of concrete structure in real time. This new vision-based crack inspection method shows great promise for the practical application.

scholarly journals A Hierarchical SVM Based Behavior Inference of Human Operators Using a Hybrid Sequence Kernel

2019 ◽  
Vol 11 (18) ◽  
pp. 4836 ◽  
Author(s):  
Jaeseok Huh ◽  
Jonghun Park ◽  
Dongmin Shin ◽  
Yerim Choi
Keyword(s):  
Support Vector Machine ◽  
Real Time ◽  
Support Vector ◽  
Inference Method ◽  
Training Environment ◽  
Log Data ◽  
Hybrid Sequence ◽  
The Hierarchical Structure ◽  
Aerial Vehicle ◽  
Human Operators

To train skilled unmanned combat aerial vehicle (UCAV) operators, it is important to establish a real-time training environment where an enemy appropriately responds to the action performed by a trainee. This can be addressed by constructing the inference method for the behavior of a UCAV operator from given simulation log data. Through this method, the virtual enemy is capable of performing actions that are highly likely to be made by an actual operator. To achieve this, we propose a hybrid sequence (HS) kernel-based hierarchical support vector machine (HSVM) for the behavior inference of a UCAV operator. Specifically, the HS kernel is designed to resolve the heterogeneity in simulation log data, and HSVM performs the behavior inference in a sequential manner considering the hierarchical structure of the behaviors of a UCAV operator. The effectiveness of the proposed method is demonstrated with the log data collected from the air-to-air combat simulator.

scholarly journals Identifying the Branch of Kiwifruit Based on Unmanned Aerial Vehicle (UAV) Images Using Deep Learning Method

Sensors ◽  
2021 ◽  
Vol 21 (13) ◽  
pp. 4442
Author(s):  
Zijie Niu ◽  
Juntao Deng ◽  
Xu Zhang ◽  
Jun Zhang ◽  
Shijia Pan ◽  
...  
Keyword(s):  
Deep Learning ◽  
Unmanned Aerial Vehicle ◽  
Semantic Segmentation ◽  
Dynamic Monitoring ◽  
Support Vector ◽  
Distribution Maps ◽  
Time Operation ◽  
Aerial Vehicle ◽  
Uav Images ◽  
Segmentation Image

It is important to obtain accurate information about kiwifruit vines to monitoring their physiological states and undertake precise orchard operations. However, because vines are small and cling to trellises, and have branches laying on the ground, numerous challenges exist in the acquisition of accurate data for kiwifruit vines. In this paper, a kiwifruit canopy distribution prediction model is proposed on the basis of low-altitude unmanned aerial vehicle (UAV) images and deep learning techniques. First, the location of the kiwifruit plants and vine distribution are extracted from high-precision images collected by UAV. The canopy gradient distribution maps with different noise reduction and distribution effects are generated by modifying the threshold and sampling size using the resampling normalization method. The results showed that the accuracies of the vine segmentation using PSPnet, support vector machine, and random forest classification were 71.2%, 85.8%, and 75.26%, respectively. However, the segmentation image obtained using depth semantic segmentation had a higher signal-to-noise ratio and was closer to the real situation. The average intersection over union of the deep semantic segmentation was more than or equal to 80% in distribution maps, whereas, in traditional machine learning, the average intersection was between 20% and 60%. This indicates the proposed model can quickly extract the vine distribution and plant position, and is thus able to perform dynamic monitoring of orchards to provide real-time operation guidance.

Sign in / Sign up

Export Citation Format

Share Document