scholarly journals Detecting Objects from Space: An Evaluation of Deep-Learning Modern Approaches

Electronics ◽  
2020 ◽  
Vol 9 (4) ◽  
pp. 583 ◽  
Author(s):  
Khang Nguyen ◽  
Nhut T. Huynh ◽  
Phat C. Nguyen ◽  
Khanh-Duy Nguyen ◽  
Nguyen D. Vo ◽  
...  
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Unmanned Aircraft ◽  
Aerial Images ◽  
Great Success ◽  
Single Shot ◽  
Convolutional Networks ◽  
Image Pyramids ◽  
Fully Convolutional Networks ◽  
Wide Range

Unmanned aircraft systems or drones enable us to record or capture many scenes from the bird’s-eye view and they have been fast deployed to a wide range of practical domains, i.e., agriculture, aerial photography, fast delivery and surveillance. Object detection task is one of the core steps in understanding videos collected from the drones. However, this task is very challenging due to the unconstrained viewpoints and low resolution of captured videos. While deep-learning modern object detectors have recently achieved great success in general benchmarks, i.e., PASCAL-VOC and MS-COCO, the robustness of these detectors on aerial images captured by drones is not well studied. In this paper, we present an evaluation of state-of-the-art deep-learning detectors including Faster R-CNN (Faster Regional CNN), RFCN (Region-based Fully Convolutional Networks), SNIPER (Scale Normalization for Image Pyramids with Efficient Resampling), Single-Shot Detector (SSD), YOLO (You Only Look Once), RetinaNet, and CenterNet for the object detection in videos captured by drones. We conduct experiments on VisDrone2019 dataset which contains 96 videos with 39,988 annotated frames and provide insights into efficient object detectors for aerial images.

scholarly journals Image-Based Plant Disease Identification by Deep Learning Meta-Architectures

Plants ◽  
2020 ◽  
Vol 9 (11) ◽  
pp. 1451
Author(s):  
Muhammad Hammad Saleem ◽  
Sapna Khanchi ◽  
Johan Potgieter ◽  
Khalid Mahmood Arif
Keyword(s):  
Deep Learning ◽  
Plant Disease ◽  
State Of The Art ◽  
Mean Average Precision ◽  
Single Shot ◽  
Average Precision ◽  
Crop Monitoring ◽  
Convolutional Networks ◽  
Fully Convolutional Networks ◽  
Agricultural Applications

The identification of plant disease is an imperative part of crop monitoring systems. Computer vision and deep learning (DL) techniques have been proven to be state-of-the-art to address various agricultural problems. This research performed the complex tasks of localization and classification of the disease in plant leaves. In this regard, three DL meta-architectures including the Single Shot MultiBox Detector (SSD), Faster Region-based Convolutional Neural Network (RCNN), and Region-based Fully Convolutional Networks (RFCN) were applied by using the TensorFlow object detection framework. All the DL models were trained/tested on a controlled environment dataset to recognize the disease in plant species. Moreover, an improvement in the mean average precision of the best-obtained deep learning architecture was attempted through different state-of-the-art deep learning optimizers. The SSD model trained with an Adam optimizer exhibited the highest mean average precision (mAP) of 73.07%. The successful identification of 26 different types of defected and 12 types of healthy leaves in a single framework proved the novelty of the work. In the future, the proposed detection methodology can also be adopted for other agricultural applications. Moreover, the generated weights can be reused for future real-time detection of plant disease in a controlled/uncontrolled environment.

scholarly journals Image-Based Iron Slag Segmentation via Graph Convolutional Networks

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Wang Long ◽  
Zheng Junfeng ◽  
Yu Hong ◽  
Ding Meng ◽  
Li Jiangyun
Keyword(s):  
Deep Learning ◽  
Contextual Information ◽  
Recognition Algorithm ◽  
Smart Manufacturing ◽  
Convolutional Network ◽  
Slag Removal ◽  
Convolutional Networks ◽  
Fully Convolutional Networks ◽  
Iron Slag ◽  
Wide Range

Slagging-off (i.e., slag removal) is an important preprocessing operation of steel-making to improve the purity of iron. Current manual-operated slag removal schemes are inefficient and labor-intensive. Automatic slagging-off is desirable but challenging as the reliable recognition of iron and slag is difficult. This work focuses on realizing an efficient and accurate recognition algorithm of iron and slag, which is conducive to realize automatic slagging-off operation. Motivated by the recent success of deep learning techniques in smart manufacturing, we introduce deep learning methods to this field for the first time. The monotonous gray value of industry images, poor image quality, and nonrigid feature of iron and slag challenge the existing fully convolutional networks (FCNs). To this end, we propose a novel spatial and feature graph convolutional network (SFGCN) module. SFGCN module can be easily inserted in FCNs to improve the reasoning ability of global contextual information, which is helpful to enhance the segmentation accuracy of small objects and isolated areas. To verify the validity of the SFGCN module, we create an industrial dataset and conduct extensive experiments. Finally, the results show that our SFGCN module brings a consistent performance boost for a wide range of FCNs. Moreover, by adopting a lightweight network as backbone, our method achieves real-time iron and slag segmentation. In the future work, we will dedicate our efforts to the weakly supervised learning for quick annotation of big data stream to improve the generalization ability of current models.

scholarly journals Design of Desktop Audiovisual Entertainment System with Deep Learning and Haptic Sensations

Symmetry ◽  
2020 ◽  
Vol 12 (10) ◽  
pp. 1718
Author(s):  
Chien-Hsing Chou ◽  
Yu-Sheng Su ◽  
Che-Ju Hsu ◽  
Kong-Chang Lee ◽  
Ping-Hsuan Han
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
User Experience ◽  
Recognition System ◽  
Scene Recognition ◽  
Single Shot ◽  
Auditory Signals ◽  
Hot Weather ◽  
Viewing Experience ◽  
At Home

In this study, we designed a four-dimensional (4D) audiovisual entertainment system called Sense. This system comprises a scene recognition system and hardware modules that provide haptic sensations for users when they watch movies and animations at home. In the scene recognition system, we used Google Cloud Vision to detect common scene elements in a video, such as fire, explosions, wind, and rain, and further determine whether the scene depicts hot weather, rain, or snow. Additionally, for animated videos, we applied deep learning with a single shot multibox detector to detect whether the animated video contained scenes of fire-related objects. The hardware module was designed to provide six types of haptic sensations set as line-symmetry to provide a better user experience. After the system considers the results of object detection via the scene recognition system, the system generates corresponding haptic sensations. The system integrates deep learning, auditory signals, and haptic sensations to provide an enhanced viewing experience.

scholarly journals Automatic Deep Learning Semantic Segmentation of Ultrasound Thyroid Cineclips using Recurrent Fully Convolutional Networks

IEEE Access ◽  
2020 ◽  
pp. 1-1
Author(s):  
Jeremy M. Webb ◽  
Duane D. Meixner ◽  
Shaheeda A. Adusei ◽  
Eric C. Polley ◽  
Mostafa Fatemi ◽  
...  
Keyword(s):  
Deep Learning ◽  
Semantic Segmentation ◽  
Convolutional Networks ◽  
Fully Convolutional Networks

scholarly journals Deep Learning-Based Object Detection for Unmanned Aerial Systems (UASs)-Based Inspections of Construction Stormwater Practices

Sensors ◽  
2021 ◽  
Vol 21 (8) ◽  
pp. 2834
Author(s):  
Billur Kazaz ◽  
Subhadipto Poddar ◽  
Saeed Arabi ◽  
Michael A. Perez ◽  
Anuj Sharma ◽  
...  
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Pollution Prevention ◽  
Aerial Images ◽  
Unmanned Aerial Systems ◽  
Construction Sites ◽  
State Regulations ◽  
Novel Approach ◽  
Model Training ◽  
Aerial Systems

Construction activities typically create large amounts of ground disturbance, which can lead to increased rates of soil erosion. Construction stormwater practices are used on active jobsites to protect downstream waterbodies from offsite sediment transport. Federal and state regulations require routine pollution prevention inspections to ensure that temporary stormwater practices are in place and performing as intended. This study addresses the existing challenges and limitations in the construction stormwater inspections and presents a unique approach for performing unmanned aerial system (UAS)-based inspections. Deep learning-based object detection principles were applied to identify and locate practices installed on active construction sites. The system integrates a post-processing stage by clustering results. The developed framework consists of data preparation with aerial inspections, model training, validation of the model, and testing for accuracy. The developed model was created from 800 aerial images and was used to detect four different types of construction stormwater practices at 100% accuracy on the Mean Average Precision (MAP) with minimal false positive detections. Results indicate that object detection could be implemented on UAS-acquired imagery as a novel approach to construction stormwater inspections and provide accurate results for site plan comparisons by rapidly detecting the quantity and location of field-installed stormwater practices.

scholarly journals BUILDING CHANGE DETECTION FROM BITEMPORAL AERIAL IMAGES USING DEEP LEARNING

2020 ◽  
Vol V-2-2020 ◽  
pp. 565-571
Author(s):  
S. Su ◽  
T. Nawata ◽  
T. Fuse
Keyword(s):  
Deep Learning ◽  
Change Detection ◽  
Urban Areas ◽  
Network Architecture ◽  
Aerial Images ◽  
Detection Accuracy ◽  
Detection Model ◽  
Wide Range ◽  
Surface Models ◽  
Data Source

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.

scholarly journals Going Deep into Remote Sensing Spatial Feature Learning

2020 ◽  
Author(s):  
Keiller Nogueira ◽  
William Robson Schwartz ◽  
Jefersson Alex Dos Santos
Keyword(s):  
Remote Sensing ◽  
Deep Learning ◽  
Feature Learning ◽  
Aerial Images ◽  
Morphological Operations ◽  
Pixel Classification ◽  
Specific Data ◽  
Convolutional Networks ◽  
Spatial Features ◽  
New Strategy

A lot of information may be extracted from the Earth’s surface through aerial images. This information may assist in myriad applications, such as urban planning, crop and forest management, disaster relief, etc. However, the process of distilling this information is strongly based on efficiently encoding the spatial features, a challenging task. Facing this, Deep Learning is able to learn specific data-driven features. This PhD thesis1 introduces deep learning into the remote sensing domain. Specifically, we tackled two main tasks, scene and pixel classification, using Deep Learning to encode spatial features over high-resolution remote sensing images. First, we proposed an architecture and analyze different strategies to exploit Convolutional Networks for image classification. Second, we introduced a network and proposed a new strategy to better exploit multi-context information in order to improve pixelwise classification. Finally, we proposed a new network based on morphological operations towards better learning of some relevant visual features.

scholarly journals Potato crop stress identification in aerial images using deep learning‐based object detection

2021 ◽  
Author(s):  
Sujata Butte ◽  
Aleksandar Vakanski ◽  
Kasia Duellman ◽  
Haotian Wang ◽  
Amin Mirkouei
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Potato Crop ◽  
Aerial Images

scholarly journals Ship Classification in High-Resolution SAR Images Using Deep Learning of Small Datasets

Sensors ◽  
2018 ◽  
Vol 18 (9) ◽  
pp. 2929 ◽  
Author(s):  
Yuanyuan Wang ◽  
Chao Wang ◽  
Hong Zhang
Keyword(s):  
Deep Learning ◽  
High Resolution ◽  
Classification Accuracy ◽  
Experimental Results ◽  
Fine Tuning ◽  
Classification Model ◽  
Great Success ◽  
Sar Images ◽  
Convolutional Networks ◽  
Ship Classification

With the capability to automatically learn discriminative features, deep learning has experienced great success in natural images but has rarely been explored for ship classification in high-resolution SAR images due to the training bottleneck caused by the small datasets. In this paper, convolutional neural networks (CNNs) are applied to ship classification by using SAR images with the small datasets. First, ship chips are constructed from high-resolution SAR images and split into training and validation datasets. Second, a ship classification model is constructed based on very deep convolutional networks (VGG). Then, VGG is pretrained via ImageNet, and fine tuning is utilized to train our model. Six scenes of COSMO-SkyMed images are used to evaluate our proposed model with regard to the classification accuracy. The experimental results reveal that (1) our proposed ship classification model trained by fine tuning achieves more than 95% average classification accuracy, even with 5-cross validation; (2) compared with other models, the ship classification model based on VGG16 achieves at least 2% higher accuracies for classification. These experimental results reveal the effectiveness of our proposed method.

Deep Learning Based Active Monitoring for Anti-collision between Vessels and Bridges

2019 ◽  
Author(s):  
Limu Chen ◽  
Ye Xia ◽  
Dexiong Pan ◽  
Chengbin Wang
Keyword(s):  
Decision Making ◽  
Deep Learning ◽  
Object Detection ◽  
Large Scale ◽  
Data Augmentation ◽  
Information Support ◽  
Single Shot ◽  
Active Monitoring ◽  
Detection Model ◽  
Comparison Results

<p>Deep-learning based navigational object detection is discussed with respect to active monitoring system for anti-collision between vessel and bridge. Motion based object detection method widely used in existing anti-collision monitoring systems is incompetent in dealing with complicated and changeable waterway for its limitations in accuracy, robustness and efficiency. The video surveillance system proposed contains six modules, including image acquisition, detection, tracking, prediction, risk evaluation and decision-making, and the detection module is discussed in detail. A vessel-exclusive dataset with tons of image samples is established for neural network training and a SSD (Single Shot MultiBox Detector) based object detection model with both universality and pertinence is generated attributing to tactics of sample filtering, data augmentation and large-scale optimization, which make it capable of stable and intelligent vessel detection. Comparison results with conventional methods indicate that the proposed deep-learning method shows remarkable advantages in robustness, accuracy, efficiency and intelligence. In-situ test is carried out at Songpu Bridge in Shanghai, and the results illustrate that the method is qualified for long-term monitoring and providing information support for further analysis and decision making.</p>

Sign in / Sign up

Export Citation Format

Share Document