Detection and Identification of Vehicles from High-Resolution Aerial Images Using Deep Learning Approaches with the Tuned Parameters

Unmanned Aerial Systems, hereafter referred to as UAS, are of great use in hazard events such as wildfire due to their ability to provide high-resolution video imagery over areas deemed too dangerous for manned aircraft and ground crews. This aerial perspective allows for identification of ground-based hazards such as spot fires and fire lines, and to communicate this information with fire fighting crews. Current technology relies on visual interpretation of UAS imagery, with little to no computer-assisted automatic detection. With the help of big labeled data and the significant increase of computing power, deep learning has seen great successes on object detection with fixed patterns, such as people and vehicles. However, little has been done for objects, such as spot fires, with amorphous and irregular shapes. Additional challenges arise when data are collected via UAS as high-resolution aerial images or videos; an ample solution must provide reasonable accuracy with low delays. In this paper, we examined 4K ( 3840 × 2160 ) videos collected by UAS from a controlled burn and created a set of labeled video sets to be shared for public use. We introduce a coarse-to-fine framework to auto-detect wildfires that are sparse, small, and irregularly-shaped. The coarse detector adaptively selects the sub-regions that are likely to contain the objects of interest while the fine detector passes only the details of the sub-regions, rather than the entire 4K region, for further scrutiny. The proposed two-phase learning therefore greatly reduced time overhead and is capable of maintaining high accuracy. Compared against the real-time one-stage object backbone of YoloV3, the proposed methods improved the mean average precision(mAP) from 0 . 29 to 0 . 67 , with an average inference speed of 7.44 frames per second. Limitations and future work are discussed with regard to the design and the experiment results.

Download Full-text

Semantic Segmentation of Urban Buildings Using a High-Resolution Network (HRNet) with Channel and Spatial Attention Gates

Remote Sensing ◽

10.3390/rs13163087 ◽

2021 ◽

Vol 13 (16) ◽

pp. 3087

Author(s):

Seonkyeong Seong ◽

Jaewan Choi

Keyword(s):

Deep Learning ◽

High Resolution ◽

Spatial Attention ◽

Semantic Segmentation ◽

Aerial Images ◽

Building Extraction ◽

Learning Models ◽

Urban Buildings

In this study, building extraction in aerial images was performed using csAG-HRNet by applying HRNet-v2 in combination with channel and spatial attention gates. HRNet-v2 consists of transition and fusion processes based on subnetworks according to various resolutions. The channel and spatial attention gates were applied in the network to efficiently learn important features. A channel attention gate assigns weights in accordance with the importance of each channel, and a spatial attention gate assigns weights in accordance with the importance of each pixel position for the entire channel. In csAG-HRNet, csAG modules consisting of a channel attention gate and a spatial attention gate were applied to each subnetwork of stage and fusion modules in the HRNet-v2 network. In experiments using two datasets, it was confirmed that csAG-HRNet could minimize false detections based on the shapes of large buildings and small nonbuilding objects compared to existing deep learning models.

Download Full-text

High-Resolution Reconstructions of Aerial Images Based On Deep Learning

Computación y Sistemas ◽

10.13053/cys-25-4-4047 ◽

2021 ◽

Vol 25 (4) ◽

Author(s):

Armando Levid Rodríguez-Santiago ◽

José Aníbal Arias-Aguilar ◽

Hiroshi Takemura ◽

Alberto Elías Petrilli-Barcelo

Keyword(s):

Deep Learning ◽

High Resolution ◽

Aerial Images

Download Full-text

Using high‐resolution aerial imagery and deep learning to detect tree spatio-temporal dynamics at the treeline

10.5194/egusphere-egu21-14548 ◽

2021 ◽

Author(s):

Mirela Beloiu ◽

Dimitris Poursanidis ◽

Samuel Hoffmann ◽

Nektarios Chrysoulakis ◽

Carl Beierkuhnlein

Keyword(s):

Deep Learning ◽

High Resolution ◽

Object Detection ◽

Temporal Dynamics ◽

Tree Density ◽

Aerial Imagery ◽

Aerial Images ◽

Remote Areas ◽

Spatio Temporal ◽

Study Sites

<p>Recent advances in deep learning techniques for object detection and the availability of high-resolution images facilitate the analysis of both temporal and spatial vegetation patterns in remote areas. High-resolution satellite imagery has been used successfully to detect trees in small areas with homogeneous rather than heterogeneous forests, in which single tree species have a strong contrast compared to their neighbors and landscape. However, no research to date has detected trees at the treeline in the remote and complex heterogeneous landscape of Greece using deep learning methods. We integrated high-resolution aerial images, climate data, and topographical characteristics to study the treeline dynamic over 70 years in the Samaria National Park on the Mediterranean island of Crete, Greece. We combined mapping techniques with deep learning approaches to detect and analyze spatio-temporal dynamics in treeline position and tree density. We use visual image interpretation to detect single trees on high-resolution aerial imagery from 1945, 2008, and 2015. Using the RGB aerial images from 2008 and 2015 we test a Convolution Neural Networks (CNN)-object detection approach (SSD) and a CNN-based segmentation technique (U-Net). Based on the mapping and deep learning approach, we have not detected a shift in treeline elevation over the last 70 years, despite warming, although tree density has increased. However, we show that CNN approach accurately detects and maps tree position and density at the treeline. We also reveal that the treeline elevation on Crete varies with topography. Treeline elevation decreases from the southern to the northern study sites. We explain these differences between study sites by the long-term interaction between topographical characteristics and meteorological factors. The study highlights the feasibility of using deep learning and high-resolution imagery as a promising technique for monitoring forests in remote areas.</p>

Download Full-text

Deep Learning Approaches on Defect Detection in High Resolution Aerial Images of Insulators

Sensors ◽

10.3390/s21041033 ◽

2021 ◽

Vol 21 (4) ◽

pp. 1033

Author(s):

Qiaodi Wen ◽

Ziqi Luo ◽

Ruitao Chen ◽

Yifan Yang ◽

Guofa Li

Keyword(s):

Neural Network ◽

Deep Learning ◽

High Resolution ◽

Convolutional Neural Network ◽

Target Detection ◽

Defect Detection ◽

Region Of Interest ◽

Aerial Images ◽

Detection Methods ◽

Complex Background

By detecting the defect location in high-resolution insulator images collected by unmanned aerial vehicle (UAV) in various environments, the occurrence of power failure can be timely detected and the caused economic loss can be reduced. However, the accuracies of existing detection methods are greatly limited by the complex background interference and small target detection. To solve this problem, two deep learning methods based on Faster R-CNN (faster region-based convolutional neural network) are proposed in this paper, namely Exact R-CNN (exact region-based convolutional neural network) and CME-CNN (cascade the mask extraction and exact region-based convolutional neural network). Firstly, we proposed an Exact R-CNN based on a series of advanced techniques including FPN (feature pyramid network), cascade regression, and GIoU (generalized intersection over union). RoI Align (region of interest align) is introduced to replace RoI pooling (region of interest pooling) to address the misalignment problem, and the depthwise separable convolution and linear bottleneck are introduced to reduce the computational burden. Secondly, a new pipeline is innovatively proposed to improve the performance of insulator defect detection, namely CME-CNN. In our proposed CME-CNN, an insulator mask image is firstly generated to eliminate the complex background by using an encoder-decoder mask extraction network, and then the Exact R-CNN is used to detect the insulator defects. The experimental results show that our proposed method can effectively detect insulator defects, and its accuracy is better than the examined mainstream target detection algorithms.

Download Full-text

Recognition of Image-Based Plant Leaf Diseases Using Deep Learning Classification Models

Nature Environment and Pollution Technology ◽

10.46488/nept.2021.v20i05.031 ◽

2021 ◽

Vol 20 (5) ◽

Author(s):

Sakshi Takkar ◽

Anuj Kakran ◽

Veerpal Kaur ◽

Manik Rakhra ◽

Manish Sharma ◽

...

Keyword(s):

Deep Learning ◽

Agricultural Land ◽

Super Resolution ◽

Modern Technology ◽

Primary Objective ◽

Plant Diseases ◽

Learning Approaches ◽

Agricultural Field ◽

Detection And Identification ◽

The Individual

Plant diseases are spread by a variety of pests, weeds, and pathogens and may have a devastating effect on agriculture, if not handled in a timely manner. Farmers face umpteen challenges from a proper water supply, untimely rain, storage facilities, and several plant diseases. Crops disease is the primary threat and it causes enormous loss to farmers in terms of production and finance. Identifying the disease from several hectares of agricultural land is a very difficult practice even with the presence of modern technology. Accurate and rapid illness prediction for early illness treatment to crops minimizes economical loss to the individual and further proves to be productive for healthy crops. Many studies use modern deep learning approaches to improve the accuracy and performance of object detection and identification systems. The suggested method notifies farmers of different agricultural illnesses, prompting them to take further essential precautions before the disease spreads to the whole agricultural field. The primary objective of this study is to detect the illnesses as soon as they begin to spread on the leaves of the plants. Super-Resolution Convolutional Neural Network (SRCNN) and Bicubic models are employed in the system to identify healthy and diseased leaves with an accuracy of 99.175 % and 99.156 % respectively.

Download Full-text

RF-based drone detection and identification using deep learning approaches: An initiative towards a large open source drone database

Future Generation Computer Systems ◽

10.1016/j.future.2019.05.007 ◽

2019 ◽

Vol 100 ◽

pp. 86-97 ◽

Cited By ~ 14

Author(s):

Mohammad F. Al-Sa’d ◽

Abdulla Al-Ali ◽

Amr Mohamed ◽

Tamer Khattab ◽

Aiman Erbad

Keyword(s):

Deep Learning ◽

Open Source ◽

Learning Approaches ◽

Detection And Identification

Download Full-text

HDFNet: Hierarchical Dynamic Fusion Network for Change Detection in Optical Aerial Images

Remote Sensing ◽

10.3390/rs13081440 ◽

2021 ◽

Vol 13 (8) ◽

pp. 1440

Author(s):

Yi Zhang ◽

Lei Fu ◽

Ying Li ◽

Yanning Zhang

Keyword(s):

Deep Learning ◽

Change Detection ◽

Aerial Images ◽

Aerial Image ◽

Sufficient Information ◽

Learning Approaches ◽

Learning Techniques ◽

Benchmark Datasets ◽

Deep Learning Features ◽

Global And Local

Accurate change detection in optical aerial images by using deep learning techniques has been attracting lots of research efforts in recent years. Correct change-detection results usually involve both global and local deep learning features. Existing deep learning approaches have achieved good performance on this task. However, under the scenarios of containing multiscale change areas within a bi-temporal image pair, existing methods still have shortcomings in adapting these change areas, such as false detection and limited completeness in detected areas. To deal with these problems, we design a hierarchical dynamic fusion network (HDFNet) to implement the optical aerial image-change detection task. Specifically, we propose a change-detection framework with hierarchical fusion strategy to provide sufficient information encouraging for change detection and introduce dynamic convolution modules to self-adaptively learn from this information. Also, we use a multilevel supervision strategy with multiscale loss functions to supervise the training process. Comprehensive experiments are conducted on two benchmark datasets, LEBEDEV and LEVIR-CD, to verify the effectiveness of the proposed method and the experimental results show that our model achieves state-of-the-art performance.

Download Full-text