scholarly journals Modified YOLOv4 for real-time Coconut Trees Detection from an Unmanned Aerial Vehicle

2021 ◽  
Author(s):  
Kshitij Nair Nair ◽  
Sibi Chakkaravarthy S ◽  
Ramya Krishna Dhulipalla ◽  
Suresh Chandra Satapathy ◽  
Abhrankash Kanungo ◽  
...  
Keyword(s):  
Real Time ◽  
Unmanned Aerial Vehicle ◽  
Simple Procedure ◽  
Field Studies ◽  
Object Localization ◽  
Aerial Images ◽  
Multiple Sources ◽  
Coconut Tree ◽  
Aerial Vehicle ◽  
Bounding Boxes

Abstract In this paper, a novel system is proposed to detect coconut trees from the images taken on an unmanned aerial vehicle (UAV). We propose a model based on the YOLOv4 Detector (CSPDarknet53) to achieve accurate prediction and fast speeds that enable real-time detection and object localization for a single object class-coconut tree. The pipeline is trained on two separate datasets of images from camera embedded UAV and images from multiple sources are fed into the multi-scale detector to predict bounding boxes and labels. A simple procedure is also proposed to enable detection on a larger scale, whereby the bigger image can be cropped into multiple single images and then fed into the detector. This model is compared statistically with the other state-of-the-art deep-learning models for object detection. After field studies and experiments on the images shot via a UAV (drone), it is proved that this system can efficiently and accurately detect coconut trees on a field at a speed of 15 frames per second when trained on 500 aerial images of the coconut trees.

scholarly journals Utilizing drone technology in the civil engineering

2019 ◽  
Vol 14 (1) ◽  
pp. 27-37
Author(s):  
Matúš Tkáč ◽  
Peter Mésároš
Keyword(s):  
Real Time ◽  
Unmanned Aerial Vehicle ◽  
Civil Engineering ◽  
Aerial Images ◽  
General Overview ◽  
Aerial Photogrammetry ◽  
Different Types ◽  
Aerial Vehicle

Abstract An unmanned aerial vehicle (UAVs), also known as drone technology, is used for different types of application in the civil engineering. Drones as a tools that increase communication between construction participants, improves site safety, uses topographic measurements of large areas, with using principles of aerial photogrammetry is possible to create buildings aerial surveying, bridges, roads, highways, saves project time and costs, etc. The use of UAVs in the civil engineering can brings many benefits; creating real-time aerial images from the building objects, overviews reveal assets and challenges, as well as the broad lay of the land, operators can share the imaging with personnel on site, in headquarters and with sub-contractors, planners can meet virtually to discuss project timing, equipment needs and challenges presented by the terrain. The aim of this contribution is to create a general overview of the use of UAVs in the civil engineering. The contribution also contains types of UAVs used for construction purposes, their advantages and also disadvantages.

scholarly journals Pemanfaatan Metode Fotogrametri untuk Pengukuran Garis Pantai dan Identifikasi Objek-Objek Tematik dengan Menggunakan Wahana Uav (Unmanned Aerial Vehicle) (Studi Kasus Pengukuran Garis Pantai di Pangkalan TNI AL Pondok Dayung

2020 ◽  
Vol 5 (1) ◽  
pp. 71-84
Author(s):  
Adhyta Harfan ◽  
Dipo Yudhatama ◽  
Imam Bachrodin
Keyword(s):  
Real Time ◽  
Unmanned Aerial Vehicle ◽  
Control Point ◽  
Ground Control ◽  
Ground Control Point ◽  
Real Time Kinematic ◽  
Aerial Vehicle

Metode Fotogrametri telah banyak digunakan dalam survei dan pemetaan. Seiring dengan kemajuan ilmu pengetahuan dan teknologi, metode fotogrametri saat ini berbasiskan pesawat tanpa awak atau yang lebih dikenal dengan UAV (Unmanned Aerial Vehicle). Kelebihan metode fotogrametri berbasiskan UAV untuk pengukuran garis pantai adalah memiliki resolusi spasial yang sangat tinggi dan dapat menjagkau daerah-daerah yang sulit dan berbahaya. Di samping itu juga dapat memberikan data foto udara terkini dengan sekala detail. Dalam penelitian ini membandingkan ketelitian horisontal antara hasil pengukuran garis pantai menggunakan metode fotogrametri berbasiskan UAV secara rektifikasi dengan GCP (Ground Control Point) maupun secara PPK (Post Processed Kinematic) dengan pengukuran garis pantai metode GNSS RTK (Real Time Kinematic). Hasil perhitungan ketelitian horisontal mengacu pada standar publikasi IHO S-44 tentang pengukuran garis pantai. Pemotretan dilakukan dengan ketinggian terbang 180 m, dengan tampalan depan dan samping 80%. Hasil perhitungan ketelitian horisontal foto udara terektifikasi 5 GCP, foto udara PPK dan foto udara PPK terektifikasi 1 GCP terhadap pengukuran garis pantai dengan metode GNSS RTK diperoleh nilai standar deviasi (σ) dan 95% selang kepercayaan (CI95%) masing-masing sebagai berikut: σ5gcp=10,989 cm dengan CI95% 16.8 cm < μ < 21.2 cm , σppk=26,066 cm dengan CI95% 26.5 cm < μ < 37 cm dan σppk1gcp=10,378 cm dengan CI95% 15.6 cm < μ < 19.8 cm. Kemudian terdapat 10 objek tematik berdasarkan Peta Laut Nomor 1 yang dapat diinterpretasi pada hasil orthomosaic foto udara.

scholarly journals Real-Time Onboard 3D State Estimation of an Unmanned Aerial Vehicle in Multi-Environments Using Multi-Sensor Data Fusion

Sensors ◽  
2020 ◽  
Vol 20 (3) ◽  
pp. 919 ◽  
Author(s):  
Hao Du ◽  
Wei Wang ◽  
Chaowen Xu ◽  
Ran Xiao ◽  
Changyin Sun
Keyword(s):  
Data Fusion ◽  
State Estimation ◽  
Real Time ◽  
Unmanned Aerial Vehicle ◽  
Position Estimation ◽  
Sensor Data ◽  
Measurement Unit ◽  
Sensor Data Fusion ◽  
Multi Sensor Data Fusion ◽  
Aerial Vehicle

The question of how to estimate the state of an unmanned aerial vehicle (UAV) in real time in multi-environments remains a challenge. Although the global navigation satellite system (GNSS) has been widely applied, drones cannot perform position estimation when a GNSS signal is not available or the GNSS is disturbed. In this paper, the problem of state estimation in multi-environments is solved by employing an Extended Kalman Filter (EKF) algorithm to fuse the data from multiple heterogeneous sensors (MHS), including an inertial measurement unit (IMU), a magnetometer, a barometer, a GNSS receiver, an optical flow sensor (OFS), Light Detection and Ranging (LiDAR), and an RGB-D camera. Finally, the robustness and effectiveness of the multi-sensor data fusion system based on the EKF algorithm are verified by field flights in unstructured, indoor, outdoor, and indoor and outdoor transition scenarios.

Fast Data Acquisition of Aerial Images Using Unmanned Aerial Vehicle System

2014 ◽  
Vol 3 (3) ◽  
pp. 162
Author(s):  
Norhadija Darwin ◽  
Anuar Ahmad
Keyword(s):  
Data Acquisition ◽  
Root Mean Square ◽  
Unmanned Aerial Vehicle ◽  
Large Scale ◽  
Digital Camera ◽  
Aerial Images ◽  
Mean Square ◽  
Fast Data Acquisition ◽  
Vehicle System ◽  
Aerial Vehicle

The present work discusses the technique and methodology of analysing the potential of fast data acquisition of aerial images using unmanned aerial vehicle system. This study utilizes UAV system for large scale mapping by using digital camera attached to the UAV. UAV is developed from the low-altitude photogrammetric mapping to perform the accuracy of the aerial photography and the resolution of the image. The Ground Control Points (GCPs) and Check Points (CPs) are established using Rapid Static techniques through GPS observation for registration purpose in photogrammetric process. The GCPs is used in the photogrammetric processes to produce photogrammetric output while the CP is employed for accuracy assessment. A Pentax Optio W90 consumer digital camera is also used in image acquisition of the aerial photograph. Besides, this study also involves image processing and map production using Erdas Imagine 8.6 software. The accuracy of the orthophoto is determined using the equation of Root Mean Square Error (RMSE). The final result from orthophoto is compared to the ground survey using total station to show the different accuracy of DEM and planimetric survey. It is discovered that root mean square errors obtained from UAV system are ± 0.510, ± 0.564 and ± 0.622 for coordinate x, y and z respectively. Hence, it can be concluded that the accuracy obtained from UAV system is achieved in sub meter. In a nutshell, UAV system has potential use for large scale mapping in field of surveying and other diversified environmental applications especially for small area which has limited time and less man power.

scholarly journals Application of Deep-Learning Methods to Bird Detection Using Unmanned Aerial Vehicle Imagery

Sensors ◽  
2019 ◽  
Vol 19 (7) ◽  
pp. 1651 ◽  
Author(s):  
Suk-Ju Hong ◽  
Yunhyeok Han ◽  
Sang-Yeon Kim ◽  
Ah-Yeong Lee ◽  
Ghiseok Kim
Keyword(s):  
Deep Learning ◽  
Unmanned Aerial Vehicle ◽  
Migratory Bird ◽  
Aerial Images ◽  
Aerial Photographs ◽  
Detection Methods ◽  
Single Shot ◽  
Convolutional Network ◽  
Aerial Vehicle ◽  
Bird Detection

Wild birds are monitored with the important objectives of identifying their habitats and estimating the size of their populations. Especially in the case of migratory bird, they are significantly recorded during specific periods of time to forecast any possible spread of animal disease such as avian influenza. This study led to the construction of deep-learning-based object-detection models with the aid of aerial photographs collected by an unmanned aerial vehicle (UAV). The dataset containing the aerial photographs includes diverse images of birds in various bird habitats and in the vicinity of lakes and on farmland. In addition, aerial images of bird decoys are captured to achieve various bird patterns and more accurate bird information. Bird detection models such as Faster Region-based Convolutional Neural Network (R-CNN), Region-based Fully Convolutional Network (R-FCN), Single Shot MultiBox Detector (SSD), Retinanet, and You Only Look Once (YOLO) were created and the performance of all models was estimated by comparing their computing speed and average precision. The test results show Faster R-CNN to be the most accurate and YOLO to be the fastest among the models. The combined results demonstrate that the use of deep-learning-based detection methods in combination with UAV aerial imagery is fairly suitable for bird detection in various environments.

Sign in / Sign up

Export Citation Format

Share Document