scholarly journals Comparison of Object Detection and Patch-Based Classification Deep Learning Models on Mid- to Late-Season Weed Detection in UAV Imagery

2020 ◽  
Vol 12 (13) ◽  
pp. 2136 ◽  
Author(s):  
Arun Narenthiran Veeranampalayam Sivakumar ◽  
Jiating Li ◽  
Stephen Scott ◽  
Eric Psota ◽  
Amit J. Jhala ◽  
...  
Keyword(s):  
Object Detection ◽  
Weed Management ◽  
Network Models ◽  
Detection Performance ◽  
Single Shot ◽  
Neural Network Models ◽  
Weed Detection ◽  
Late Season ◽  
Detection Model ◽  
Growing Seasons

Mid- to late-season weeds that escape from the routine early-season weed management threaten agricultural production by creating a large number of seeds for several future growing seasons. Rapid and accurate detection of weed patches in field is the first step of site-specific weed management. In this study, object detection-based convolutional neural network models were trained and evaluated over low-altitude unmanned aerial vehicle (UAV) imagery for mid- to late-season weed detection in soybean fields. The performance of two object detection models, Faster RCNN and the Single Shot Detector (SSD), were evaluated and compared in terms of weed detection performance using mean Intersection over Union (IoU) and inference speed. It was found that the Faster RCNN model with 200 box proposals had similar good weed detection performance to the SSD model in terms of precision, recall, f1 score, and IoU, as well as a similar inference time. The precision, recall, f1 score and IoU were 0.65, 0.68, 0.66 and 0.85 for Faster RCNN with 200 proposals, and 0.66, 0.68, 0.67 and 0.84 for SSD, respectively. However, the optimal confidence threshold of the SSD model was found to be much lower than that of the Faster RCNN model, which indicated that SSD might have lower generalization performance than Faster RCNN for mid- to late-season weed detection in soybean fields using UAV imagery. The performance of the object detection model was also compared with patch-based CNN model. The Faster RCNN model yielded a better weed detection performance than the patch-based CNN with and without overlap. The inference time of Faster RCNN was similar to patch-based CNN without overlap, but significantly less than patch-based CNN with overlap. Hence, Faster RCNN was found to be the best model in terms of weed detection performance and inference time among the different models compared in this study. This work is important in understanding the potential and identifying the algorithms for an on-farm, near real-time weed detection and management.

scholarly journals Fast COVID-19 Detection of Chest X-Ray Images Using Single Shot Detection MobileNet Convolutional Neural Networks

2021 ◽  
Vol 56 (2) ◽  
pp. 235-248
Author(s):  
Fatchul Arifin ◽  
Herjuna Artanto ◽  
Nurhasanah ◽  
Teddy Surya Gunawan
Keyword(s):  
Neural Network ◽  
Detection System ◽  
Network Models ◽  
Single Shot ◽  
Neural Network Models ◽  
X Ray ◽  
Detection Model ◽  
Shot Detection ◽  
Average Accuracy ◽  
Chest X Ray

COVID-19 is a new disease with a very rapid and tremendous spread. The most important thing needed now is a COVID-19 early detection system that is fast, easy to use, portable, and affordable. Various studies on desktop-based detection using Convolutional Neural Networks have been successfully conducted. However, no research has yet applied mobile-based detection, which requires low computational cost. Therefore, this research aims to produce a COVID-19 early detection system based on chest X-ray images using Convolutional Neural Network models to be deployed in mobile applications. It is expected that the proposed Convolutional Neural Network models can detect COVID-19 quickly, economically, and accurately. The used architecture is MobileNet's Single Shot Detection. The advantage of the Single Shot Detection MobileNet models is that they are lightweight to be applied to mobile-based devices. Therefore, these two versions will also be tested, which one is better. Both models have successfully detected COVID-19, normal, and viral pneumonia conditions with an average overall accuracy of 93.24% based on the test results. The Single Shot Detection MobileNet V1 model can detect COVID-19 with an average accuracy of 83.7%, while the Single Shot Detection MobileNet V2 Single Shot Detection model can detect COVID-19 with an average accuracy of 87.5%. Based on the research conducted, it can be concluded that the approach to detecting chest X-rays of COVID-19 can be detected using the MobileNet Single Shot Detection model. Besides, the V2 model shows better performance than the V1. Therefore, this model can be applied to increase the speed and affordability of COVID-19 detection.

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>

scholarly journals Extracting Representative Images of Tourist Attractions from Flickr by Combining an Improved Cluster Method and Multiple Deep Learning Models

2020 ◽  
Vol 9 (2) ◽  
pp. 81 ◽  
Author(s):  
Shanshan Han ◽  
Fu Ren ◽  
Qingyun Du ◽  
Dawei Gui
Keyword(s):  
Real Life ◽  
Image Data ◽  
Network Models ◽  
Cluster Method ◽  
Single Shot ◽  
Large Area ◽  
Neural Network Models ◽  
Tourist Attractions ◽  
Geotagged Photos ◽  
Filter Noise

Extracting representative images of tourist attractions from geotagged photos is beneficial to many fields in tourist management, such as applications in touristic information systems. This task usually begins with clustering to extract tourist attractions from raw coordinates in geotagged photos. However, most existing cluster methods are limited in the accuracy and granularity of the places of interest, as well as in detecting distinct tags, due to its primary consideration of spatial relationships. After clustering, the challenge still exists for the task of extracting representative images within the geotagged base image data, because of the existence of noisy photos occupied by a large area proportion of humans and unrelated objects. In this paper, we propose a framework containing an improved cluster method and multiple neural network models to extract representative images of tourist attractions. We first propose a novel time- and user-constrained density-joinable cluster method (TU-DJ-Cluster), specific to photos with similar geotags to detect place-relevant tags. Then we merge and extend the clusters according to the similarity between pairs of tag embeddings, as trained from Word2Vec. Based on the clustering result, we filter noise images with Multilayer Perceptron and a single-shot multibox detector model, and further select representative images with the deep ranking model. We select Beijing as the study area. The quantitative and qualitative analysis, as well as the questionnaire results obtained from real-life tourists, demonstrate the effectiveness of this framework.

scholarly journals Small-Object Detection in Remote Sensing Images with End-to-End Edge-Enhanced GAN and Object Detector Network

2020 ◽  
Author(s):  
Jakaria Rabbi ◽  
Nilanjan Ray ◽  
Matthias Schubert ◽  
Subir Chowdhury ◽  
Dennis Chao
Keyword(s):  
Remote Sensing ◽  
Object Detection ◽  
Super Resolution ◽  
Detection Performance ◽  
Superior Performance ◽  
Single Shot ◽  
Small Object ◽  
Remote Sensing Images ◽  
Low Resolution ◽  
End To End

The detection performance of small objects in remote sensing images is not satisfactory compared to large objects, especially in low-resolution and noisy images. A generative adversarial network (GAN)-based model called enhanced super-resolution GAN (ESRGAN) shows remarkable image enhancement performance, but reconstructed images miss high-frequency edge information. Therefore, object detection performance degrades for the small objects on recovered noisy and low-resolution remote sensing images. Inspired by the success of edge enhanced GAN (EEGAN) and ESRGAN, we apply a new edge-enhanced super-resolution GAN (EESRGAN) to improve the image quality of remote sensing images and used different detector networks in an end-to-end manner where detector loss is backpropagated into the EESRGAN to improve the detection performance. We propose an architecture with three components: ESRGAN, Edge Enhancement Network (EEN), and Detection network. We use residual-in-residual dense blocks (RRDB) for both the GAN and EEN, and for the detector network, we use the faster region-based convolutional network (FRCNN) (two-stage detector) and single-shot multi-box detector (SSD) (one stage detector). Extensive experiments on car overhead with context and oil and gas storage tank (created by us) data sets show superior performance of our method compared to the standalone state-of-the-art object detectors.

scholarly journals Lightweight Feature Enhancement Network for Single-Shot Object Detection

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1066
Author(s):  
Peng Jia ◽  
Fuxiang Liu
Keyword(s):  
Receptive Field ◽  
Object Detection ◽  
Real Time ◽  
Detection System ◽  
Field Enhancement ◽  
Detection Performance ◽  
Feature Representation ◽  
Data Sets ◽  
Single Shot ◽  
Enhancement Method

At present, the one-stage detector based on the lightweight model can achieve real-time speed, but the detection performance is challenging. To enhance the discriminability and robustness of the model extraction features and improve the detector’s detection performance for small objects, we propose two modules in this work. First, we propose a receptive field enhancement method, referred to as adaptive receptive field fusion (ARFF). It enhances the model’s feature representation ability by adaptively learning the fusion weights of different receptive field branches in the receptive field module. Then, we propose an enhanced up-sampling (EU) module to reduce the information loss caused by up-sampling on the feature map. Finally, we assemble ARFF and EU modules on top of YOLO v3 to build a real-time, high-precision and lightweight object detection system referred to as the ARFF-EU network. We achieve a state-of-the-art speed and accuracy trade-off on both the Pascal VOC and MS COCO data sets, reporting 83.6% AP at 37.5 FPS and 42.5% AP at 33.7 FPS, respectively. The experimental results show that our proposed ARFF and EU modules improve the detection performance of the ARFF-EU network and achieve the development of advanced, very deep detectors while maintaining real-time speed.

scholarly journals AF-EMS Detector: Improve the Multi-Scale Detection Performance of the Anchor-Free Detector

2021 ◽  
Vol 13 (2) ◽  
pp. 160
Author(s):  
Jiangqiao Yan ◽  
Liangjin Zhao ◽  
Wenhui Diao ◽  
Hongqi Wang ◽  
Xian Sun
Keyword(s):  
Remote Sensing ◽  
Object Detection ◽  
Feature Fusion ◽  
Detection Algorithm ◽  
Detection Performance ◽  
Natural Image ◽  
Detection Model ◽  
Multi Scale ◽  
Feature Pyramid ◽  
Scale Detection

As a precursor step for computer vision algorithms, object detection plays an important role in various practical application scenarios. With the objects to be detected becoming more complex, the problem of multi-scale object detection has attracted more and more attention, especially in the field of remote sensing detection. Early convolutional neural network detection algorithms are mostly based on artificially preset anchor-boxes to divide different regions in the image, and then obtain the prior position of the target. However, the anchor box is difficult to set reasonably and will cause a large amount of computational redundancy, which affects the generality of the detection model obtained under fixed parameters. In the past two years, anchor-free detection algorithm has achieved remarkable development in the field of detection on natural image. However, there is no sufficient research on how to deal with multi-scale detection more effectively in anchor-free framework and use these detectors on remote sensing images. In this paper, we propose a specific-attention Feature Pyramid Network (FPN) module, which is able to generate a feature pyramid, basing on the characteristics of objects with various sizes. In addition, this pyramid suits multi-scale object detection better. Besides, a scale-aware detection head is proposed which contains a multi-receptive feature fusion module and a size-based feature compensation module. The new anchor-free detector can obtain a more effective multi-scale feature expression. Experiments on challenging datasets show that our approach performs favorably against other methods in terms of the multi-scale object detection performance.

scholarly journals Anomaly Detection Using XGBoost Ensemble of Deep Neural Network Models

2021 ◽  
Vol 21 (3) ◽  
pp. 175-188
Author(s):  
Sumaiya Thaseen Ikram ◽  
Aswani Kumar Cherukuri ◽  
Babu Poorva ◽  
Pamidi Sai Ushasree ◽  
Yishuo Zhang ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Anomaly Detection ◽  
Deep Neural Network ◽  
Short Term Memory ◽  
Network Models ◽  
Neural Network Models ◽  
Detection Model ◽  
Detection Systems ◽  
Learning Techniques

Abstract Intrusion Detection Systems (IDSs) utilise deep learning techniques to identify intrusions with maximum accuracy and reduce false alarm rates. The feature extraction is also automated in these techniques. In this paper, an ensemble of different Deep Neural Network (DNN) models like MultiLayer Perceptron (MLP), BackPropagation Network (BPN) and Long Short Term Memory (LSTM) are stacked to build a robust anomaly detection model. The performance of the ensemble model is analysed on different datasets, namely UNSW-NB15 and a campus generated dataset named VIT_SPARC20. Other types of traffic, namely unencrypted normal traffic, normal encrypted traffic, encrypted and unencrypted malicious traffic, are captured in the VIT_SPARC20 dataset. Encrypted normal and malicious traffic of VIT_SPARC20 is categorised by the deep learning models without decrypting its contents, thus preserving the confidentiality and integrity of the data transmitted. XGBoost integrates the results of each deep learning model to achieve higher accuracy. From experimental analysis, it is inferred that UNSW_ NB results in a maximal accuracy of 99.5%. The performance of VIT_SPARC20 in terms of accuracy, precision and recall are 99.4%. 98% and 97%, respectively.

scholarly journals A Decisive Object Detection using Deep Learning Techniques

2019 ◽  
Vol 9 (1S) ◽  
pp. 414-417
Keyword(s):  
Image Processing ◽  
Object Detection ◽  
The Other ◽  
Single Shot ◽  
Detection Techniques ◽  
Image Processing Techniques ◽  
Detection Model ◽  
Learning Techniques ◽  
Processing Techniques ◽  
Speed And Accuracy

Object detection is one of the essential features of computer vision and image processing techniques. In today's world, the computer can replicate or outperform the operation that a human can do. One such thing is object detection, and In the case of it, the machines must be trained in such a way that it can recognize the object equivalent to the human does with maximum accuracy. Several object detection techniques are used to train the machine to detect the objects. Some of the most common object detection techniques are R-CNN, Fast R-CNN, Faster R-CNN) Single Shot MultiBox Detector (SSD), and You Only Look Once(YOLO),. Each of these techniques has a different way of approach and accuracy of detecting the objects in real-time. These techniques are differentiated based on their performances, i.e., speed and accuracy. Some techniques may be very accurate in detecting the objects but may lack in the time taken for detecting the objects, whereas, on the other hand, some techniques may be very fast in figuring out the objects but not with greater accuracy. We have trained an object detection model based on the YOLO technique which gave the best performance out of all other existing techniques, though the accuracy of the model is less, the speed of detection is extremely high. So based on our research we have figured out the best performance object detection techniques and also the most accurate technique. A well-trained object detection model must be very optimistic in terms of their speed and accuracy.

scholarly journals Small-Object Detection in Remote Sensing Images with End-to-End Edge-Enhanced GAN and Object Detector Network

2020 ◽  
Author(s):  
Jakaria Rabbi ◽  
Nilanjan Ray ◽  
Matthias Schubert ◽  
Subir Chowdhury ◽  
Dennis Chao
Keyword(s):  
Remote Sensing ◽  
Object Detection ◽  
Super Resolution ◽  
Detection Performance ◽  
Superior Performance ◽  
Single Shot ◽  
Small Object ◽  
Remote Sensing Images ◽  
Low Resolution ◽  
End To End

The detection performance of small objects in remote sensing images is not satisfactory compared to large objects, especially in low-resolution and noisy images. A generative adversarial network (GAN)-based model called enhanced super-resolution GAN (ESRGAN) shows remarkable image enhancement performance, but reconstructed images miss high-frequency edge information. Therefore, object detection performance degrades for small objects on recovered noisy and low-resolution remote sensing images. Inspired by the success of edge enhanced GAN (EEGAN) and ESRGAN, we apply a new edge-enhanced super-resolution GAN (EESRGAN) to improve the image quality of remote sensing images and use different detector networks in an end-to-end manner where detector loss is backpropagated into the EESRGAN to improve the detection performance. We propose an architecture with three components: ESRGAN, Edge Enhancement Network (EEN), and Detection network. We use residual-in-residual dense blocks (RRDB) for both the ESRGAN and EEN, and for the detector network, we use the faster region-based convolutional network (FRCNN) (two-stage detector) and single-shot multi-box detector (SSD) (one stage detector). Extensive experiments on a public (car overhead with context) and a self-assembled (oil and gas storage tank) satellite dataset show superior performance of our method compared to the standalone state-of-the-art object detectors.

Sign in / Sign up

Export Citation Format

Share Document