scholarly journals A Comprehensive Multi-Modal Framework for Plant Health Monitoring

2021 ◽  
Vol 9 (VII) ◽  
pp. 1793-1795
Author(s):  
Mohammad Hanan Bhat
Keyword(s):  
Deep Learning ◽  
Data Collection ◽  
Data Storage ◽  
Health Monitoring ◽  
Single Point ◽  
Research Work ◽  
Plant Health ◽  
Cloud Infrastructure ◽  
Image Dataset ◽  
End To End

: Plant health monitoring has been a significant field of research since a very long time. The scope of this research work conducted lies in the vast domain of plant pathology with its applications extending in the field of agriculture production monitoring to forest health monitoring. It deals with the data collection techniques based on IOT, pre-processing and post-processing of Image dataset and identification of disease using deep learning model. Therefore, providing a multi-modal end-to-end approach for plant health monitoring. This paper reviews the various methods used for monitoring plant health remotely in a non-invasive manner. An end-to-end low cost framework has been proposed for monitoring plant health by using IOT based data collection methods and cloud computing for a single-point-of-contact for the data storage and processing. The cloud agent gateway connects the devices and collects the data from sensors to ensure a single source of truth. Further, the deep learning computational infrastructure provided by the public cloud infrastructure is exploited to train the image dataset and derive the plant health status

scholarly journals Dynamic Resource Allocation and Memory Management using Deep Convolutional Neural Network

2019 ◽  
Vol 9 (2) ◽  
pp. 608-612
Keyword(s):  
Neural Network ◽  
Resource Allocation ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Data Storage ◽  
Memory Management ◽  
Research Work ◽  
Dynamic Resource Allocation ◽  
Mobile Platform ◽  
Dynamic Resource

Memory management is very essential task for large-scale storage systems; in mobile platform generate storage errors due to insufficient memory as well as additional task overhead. Many existing systems have illustrated different solution for such issues, like load balancing and load rebalancing. Different unusable applications which are already installed in mobile platform user never access frequently but it allocates some memory space on hard device storage. In the proposed research work we describe dynamic resource allocation for mobile platforms using deep learning approach. In Real world mobile systems users may install different kind of applications which required ad-hoc basis. Such applications may be affect to execution performance of system as well space complexity, sometime they also affect another runnable applications performance. To eliminate of such issues, we carried out an approach to allocate runtime resources for data storage for mobile platform. When system connected with cloud data server it store complete file system on remote Virtual Machine (VM) and whenever a single application required which immediately install beginning as remote server to local device. For developed of proposed system we implemented deep learning base Convolutional Neural Network (CNN), algorithm has used with tensorflow environment which reduces the time complexity for data storage as well as extraction respectively.

scholarly journals Deep Learning Based Object Recognition Using Physically-Realistic Synthetic Depth Scenes

2019 ◽  
Vol 1 (3) ◽  
pp. 883-903 ◽  
Author(s):  
Daulet Baimukashev ◽  
Alikhan Zhilisbayev ◽  
Askat Kuzdeuov ◽  
Artemiy Oleinikov ◽  
Denis Fadeyev ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Object Recognition ◽  
Data Collection ◽  
Object Detection ◽  
Depth Image ◽  
Cluttered Environment ◽  
Depth Data ◽  
Depth Images ◽  
Image Dataset

Recognizing objects and estimating their poses have a wide range of application in robotics. For instance, to grasp objects, robots need the position and orientation of objects in 3D. The task becomes challenging in a cluttered environment with different types of objects. A popular approach to tackle this problem is to utilize a deep neural network for object recognition. However, deep learning-based object detection in cluttered environments requires a substantial amount of data. Collection of these data requires time and extensive human labor for manual labeling. In this study, our objective was the development and validation of a deep object recognition framework using a synthetic depth image dataset. We synthetically generated a depth image dataset of 22 objects randomly placed in a 0.5 m × 0.5 m × 0.1 m box, and automatically labeled all objects with an occlusion rate below 70%. Faster Region Convolutional Neural Network (R-CNN) architecture was adopted for training using a dataset of 800,000 synthetic depth images, and its performance was tested on a real-world depth image dataset consisting of 2000 samples. Deep object recognizer has 40.96% detection accuracy on the real depth images and 93.5% on the synthetic depth images. Training the deep learning model with noise-added synthetic images improves the recognition accuracy for real images to 46.3%. The object detection framework can be trained on synthetically generated depth data, and then employed for object recognition on the real depth data in a cluttered environment. Synthetic depth data-based deep object detection has the potential to substantially decrease the time and human effort required for the extensive data collection and labeling.

e-Nanoflex Sensor System: Smartphone-Based Roaming Health Monitor

2011 ◽  
Vol 2 (1) ◽  
Author(s):  
Vijay K. Varadan ◽  
Prashanth S. Kumar ◽  
Sechang Oh ◽  
Hyeokjun Kwon ◽  
Pratyush Rai ◽  
...  
Keyword(s):  
Data Storage ◽  
Data Dissemination ◽  
Point Of Care ◽  
Low Cost ◽  
Flexible Substrate ◽  
Diagnosis And Treatment ◽  
Machine Learning Techniques ◽  
Market Demand ◽  
Cloud Infrastructure ◽  
End To End

The growing need and market demand for point of care (POC) systems to improve patient’s quality of life are driving the development of wireless nanotechnology based smart systems for diagnosis and treatment of various chronic and life threatening diseases. POC diagnostics for neurological, metabolic, and cardiovascular disorders require constant long term untethered monitoring of individuals. Given the uncertainty associated with location and time at which immediate diagnosis and treatment may be required, constant vigilance and monitoring are the only practical solutions. What is needed is for a remote cyber-enabled health care smart system incorporating novel ideas from nanotechnology, low power embedded systems, wireless networking, and cloud computing to fundamentally advance. To meet this goal, we present e-Nanoflex platform, which is capable of monitoring patient health wherever they may be and communicating the data in real time to a physician or a hospital. Unlike state-of-the-art systems that are either local sensor systems or rely on custom relaying devices, e-Nanoflex is a highly nonintrusive and inexpensive end-to-end cyber-physical system. Using nanostructured sensors, e-Nanoflex provides nearly invisible monitoring of physiological conditions. It relies on smartphones to filter, compress, and relay geo-tagged data. Further, it ties to a backend cloud infrastructure for data storage, data dissemination, and abnormality detection using machine learning techniques. e-Nanoflex is a complete end-to-end system for physiological sensing and geo-tagged data dissemination to hospitals and caregivers. It is intended as a basic platform that can support any nanostructure based flexible sensor to monitor a variety of conditions such as body temperature, respiration air flow, oxygen consumption, bioelectric signals, pulse oximetry, muscle activity, and neural activity. Additionally, to address the cost of manufacturing sensors, e-Nanoflex uses a low cost production technique based on roll to roll gravure printing. We show the efficacy of our platform through a case study that involves acquiring electrocardiogram signals using gold nano-electrodes fabricated on a flexible substrate.

Life Cycle Development of Successful Women-owned Enterprises

2020 ◽  
Vol 3 (4) ◽  
pp. 142-152
Author(s):  
Mohammad Waliul Hasanat ◽  
Kamna Anum ◽  
Ashikul Hoque ◽  
Mahmud Hamid ◽  
Sandy Francis Peris ◽  
...  
Keyword(s):  
Life Cycle ◽  
Data Collection ◽  
Research Work ◽  
Primary Data ◽  
Small Scale ◽  
Data Collection Process ◽  
Theoretical Finding ◽  
Successful Women ◽  
Two Phases ◽  
Life Cycle Development

In developing countries, the role of women in the business sector is continuously improving. As a result, female enterprises have also been encouraged in Pakistan. This study is based on life cycle development phases from which women-owned enterprises have to go through in order to become successful. As a primary data source, face-to-face interviews with owners of successful women-owned enterprises were preferred. The data collection process was divided into two phases i.e. Phase-I and Phase-II. After data collection, qualitative analysis has been performed using NVIVO. Findings provide both generic and specific factors involved in life cycle development of women-owned enterprises. This study provides a detailed view of life cycle development model followed by successful women enterprises. The outcome of this research work is a theoretical finding which can be utilized by entrepreneurs owning small scale enterprises to improve their level of performance. Findings can also be helpful for potentially talented women interested in setting up their own business.

QoS aware Multi-Path Routing using Link Scheduling Algorithm

2019 ◽  
Vol 09 ◽  
Author(s):  
Satyasrikanth Palle ◽  
Shivashankar
Keyword(s):  
Cellular Networks ◽  
Scheduling Algorithm ◽  
Research Work ◽  
Scheduling Algorithms ◽  
Turnaround Time ◽  
Link Scheduling ◽  
Spatial Reuse ◽  
Mobile Cellular ◽  
Simulation Results ◽  
End To End

Objective: The demand for Cellular based multimedia services is growing day by day, in order to fulfill such demand the present day cellular networks needs to be upgraded to support excessive capacity calls along with high data accessibility. Analysis of traffic and huge network size could become very challenging issue for the network operators for scheduling the available bandwidth between different users. In the proposed work a novel QoS Aware Multi Path scheduling algorithm for smooth CAC in wireless mobile networks. The performance of the proposed algorithm is assessed and compared with existing scheduling algorithms. The simulation results show that the proposed algorithm outperforms existing CAC algorithms in terms of throughput and delay. The CAC algorithm with scheduling increases end-to-end throughput and decreases end-to-end delay. Methods: The key idea to implement the proposed research work is to adopt spatial reuse concept of wireless sensor networks to mobile cellular networks. Spatial reusability enhances channel reuse when the node pairs are far away and distant. When Src and node b are communicating with each other, the other nodes in the discovered path should be idle without utilizing the channel. Instead the other nodes are able to communicate parallelly the end-to-end throughput can be improved with acceptable delay. Incorporating link scheduling algorithms to this key concept further enhances the end-to-end throughput with in the turnaround time. So, in this research work we have applied spatial reuse concept along with link scheduling algorithm to enhance end-to-end throughput with in turnaround time. The proposed algorithm not only ensures that a connection gets the required bandwidth at each mobile node on its way by scheduling required slots to meet the QoS requirements. By considering the bandwidth requirement of the mobile connections, the CAC module at the BS not only considers the bandwidth requirement but also conforming the constrains of system dealy and jitter are met. Result: To verify the feasibility and effectiveness of our proposed work, with respect to scheduling the simulation results clearly shows the throughput improvement with Call Admission Control. The number of dropped calls is significantly less and successful calls are more with CAC. The percentage of dropped calls is reduced by 9 % and successful calls are improved by 91%. The simulation is also conducted on time constraint and ratio of dropped calls are shown. The total time taken to forward the packets and the ration of dropped calls is less when compared to non CAC. On a whole the CAC with scheduling algorithms out performs existing scheduling algorithms. Conclusion: In this research work we have proposed a novel QoS aware scheduling algorithm that provides QoS in Wireless Cellular Networks using Call Admission Control (CAC). The simulation results show that the end-to-end throughput has been increased by 91% when CAC is used. The proposed algorithm is also compared with existing link scheduling algorithms. The results reveal that CAC with scheduling algorithm can be used in Mobile Cellular Networks in order to reduce packet drop ratio. The algorithm is also used to send the packets within acceptable delay.

scholarly journals Matching Large Baseline Oblique Stereo Images Using an End-to-End Convolutional Neural Network

2021 ◽  
Vol 13 (2) ◽  
pp. 274
Author(s):  
Guobiao Yao ◽  
Alper Yilmaz ◽  
Li Zhang ◽  
Fei Meng ◽  
Haibin Ai ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Stereo Matching ◽  
Least Square ◽  
Affine Invariant ◽  
Stereo Images ◽  
Distance Ratio ◽  
Matching Algorithm ◽  
End To End

The available stereo matching algorithms produce large number of false positive matches or only produce a few true-positives across oblique stereo images with large baseline. This undesired result happens due to the complex perspective deformation and radiometric distortion across the images. To address this problem, we propose a novel affine invariant feature matching algorithm with subpixel accuracy based on an end-to-end convolutional neural network (CNN). In our method, we adopt and modify a Hessian affine network, which we refer to as IHesAffNet, to obtain affine invariant Hessian regions using deep learning framework. To improve the correlation between corresponding features, we introduce an empirical weighted loss function (EWLF) based on the negative samples using K nearest neighbors, and then generate deep learning-based descriptors with high discrimination that is realized with our multiple hard network structure (MTHardNets). Following this step, the conjugate features are produced by using the Euclidean distance ratio as the matching metric, and the accuracy of matches are optimized through the deep learning transform based least square matching (DLT-LSM). Finally, experiments on Large baseline oblique stereo images acquired by ground close-range and unmanned aerial vehicle (UAV) verify the effectiveness of the proposed approach, and comprehensive comparisons demonstrate that our matching algorithm outperforms the state-of-art methods in terms of accuracy, distribution and correct ratio. The main contributions of this article are: (i) our proposed MTHardNets can generate high quality descriptors; and (ii) the IHesAffNet can produce substantial affine invariant corresponding features with reliable transform parameters.

scholarly journals RDD2020: An annotated image dataset for Automatic Road Damage Detection using Deep Learning

Data in Brief ◽  
2021 ◽  
pp. 107133
Author(s):  
Deeksha Arya ◽  
Hiroya Maeda ◽  
Sanjay Kumar Ghosh ◽  
Durga Toshniwal ◽  
Yoshihide Sekimoto
Keyword(s):  
Deep Learning ◽  
Damage Detection ◽  
Image Dataset

scholarly journals An Investigation into the Application of Deep Learning in the Detection and Mitigation of DDOS Attack on SDN Controllers

Technologies ◽  
2021 ◽  
Vol 9 (1) ◽  
pp. 14
Author(s):  
James Dzisi Gadze ◽  
Akua Acheampomaa Bamfo-Asante ◽  
Justice Owusu Agyemang ◽  
Henry Nunoo-Mensah ◽  
Kwasi Adu-Boahen Opare
Keyword(s):  
Deep Learning ◽  
Network Architecture ◽  
Learning Algorithm ◽  
Single Point ◽  
Denial Of Service ◽  
Specific Work ◽  
Learning Models ◽  
Ddos Attack ◽  
Deep Learning Algorithm ◽  
Proposed Model

Software-Defined Networking (SDN) is a new paradigm that revolutionizes the idea of a software-driven network through the separation of control and data planes. It addresses the problems of traditional network architecture. Nevertheless, this brilliant architecture is exposed to several security threats, e.g., the distributed denial of service (DDoS) attack, which is hard to contain in such software-based networks. The concept of a centralized controller in SDN makes it a single point of attack as well as a single point of failure. In this paper, deep learning-based models, long-short term memory (LSTM) and convolutional neural network (CNN), are investigated. It illustrates their possibility and efficiency in being used in detecting and mitigating DDoS attack. The paper focuses on TCP, UDP, and ICMP flood attacks that target the controller. The performance of the models was evaluated based on the accuracy, recall, and true negative rate. We compared the performance of the deep learning models with classical machine learning models. We further provide details on the time taken to detect and mitigate the attack. Our results show that RNN LSTM is a viable deep learning algorithm that can be applied in the detection and mitigation of DDoS in the SDN controller. Our proposed model produced an accuracy of 89.63%, which outperformed linear-based models such as SVM (86.85%) and Naive Bayes (82.61%). Although KNN, which is a linear-based model, outperformed our proposed model (achieving an accuracy of 99.4%), our proposed model provides a good trade-off between precision and recall, which makes it suitable for DDoS classification. In addition, it was realized that the split ratio of the training and testing datasets can give different results in the performance of a deep learning algorithm used in a specific work. The model achieved the best performance when a split of 70/30 was used in comparison to 80/20 and 60/40 split ratios.

Sign in / Sign up

Export Citation Format

Share Document