scholarly journals A Survey on Abandoned Objects Detection from CCTV Surveillance

2020 ◽  
Vol 5 (2) ◽  
pp. 105-118
Author(s):  
Saluky Saluky
Keyword(s):  
Image Data ◽  
Automatic Monitoring ◽  
Visual Surveillance ◽  
Surveillance Systems ◽  
Surveillance Video ◽  
Abandoned Object ◽  
Detection Of Objects ◽  
The Government ◽  
Cctv Surveillance ◽  
Number Of Publications

Computer Vision is an important and challenging area of ​​research in image processing applied to analytical video. Image data comes from CCTV surveillance which is spread in public places owned by the government, private sector and the public. Supervision is carried out to monitor anomalies in the surrounding environment such as abandoned objects, crowds, theft and others. An abandoned object is one of the anomalies that is important to monitor because it can be categorized as a danger and can also prevent theft of the object left behind, therefore automatic monitoring is needed to prevent adverse events from occurring. In the last decade, a number of publications have been presented in the field of intelligent visual surveillance to detect abandoned objects (AOD). In this paper, we present a state-of-the-art showing the overall progress of the detection of objects that were abandoned or removed from surveillance video in recent years. We include a brief introduction to the detection of abandoned objects with their problems and challenges. The aim of this paper is to provide a review of the literature in the field of recognition of abandoned objects of visual surveillance systems with a general framework for researchers in this field.

scholarly journals An Improved Deep Learning Model for Plant Disease Detection

2020 ◽  
Vol 8 (6) ◽  
pp. 5389-5392
Keyword(s):  
Image Data ◽  
Plant Diseases ◽  
Data Generation ◽  
Surveillance Video ◽  
Agricultural Field ◽  
Data Set ◽  
Huge Data ◽  
Detection Of Objects ◽  
Plant Image ◽  
Model Training

In current era, Deep Convolution Neural Networks (DCNNs) are desperately improved localization, identification and detection of objects. Recent days, Big data is evolved which leads huge data generation through modern tools like surveillance video cameras. In this paper, we have focused on plant data images in agricultural field. Agriculture is one of major living source in India. To increase the yield by preventing diseases and detection of diseases place major role in agriculture domain. By using Improved and customized DCNN model (improved-detect), We trained plantdoc and plant village datasets. Mainly we used Tomato, Corn and potato plant for model training and testing. we have experimented on plant image data set-tomato leaves both healthy and diseased ones. Experimental results are compared with state of the architectures like Mobile Net, Dark Net-19, ResNet-101and proposed model out PERFORMS in location and detection of plant diseases. obtains best results in computation and accuracy. In the below results sections, we have presented the results with suitable models.

CVSS

2019 ◽  
pp. 19-32
Author(s):  
Lei Zhou ◽  
Wei Qi Yan ◽  
Yun Shu ◽  
Jian Yu
Keyword(s):  
Cloud Computing ◽  
Real Time ◽  
Virtual Machine ◽  
Surveillance System ◽  
Video Stream ◽  
Visual Surveillance ◽  
Video Transcoding ◽  
Surveillance Systems ◽  
Surveillance Video ◽  
And Storage

A large amount of surveillance videos and images need sufficient storage. In this article, an architecture of cloud-based surveillance systems and its modules will be designed, the Cloud-based Visual Surveillance System (CVSS) will be implemented on a private cloud using a Virtual Machine (VM). The users are able to link their cameras to the CVSS system so that the goal of this design can be achieved. The authors' CVSS system is able to push notification messages of captured videos to receivers, and their users could receive a surveillance video along with its events. The CVSS system fully makes use of the merits of cloud computing, which make it more advanced as stated in the evaluation section of this article. The contributions of this article are to be implemented in the CVSS system with: (1) video stream input, (2) intelligent visual surveillance, (3) real-time video transcoding and storage, (4) message pushing and media streaming output.

CVSS

2018 ◽  
Vol 10 (1) ◽  
pp. 79-91 ◽  
Author(s):  
Lei Zhou ◽  
Wei Qi Yan ◽  
Yun Shu ◽  
Jian Yu
Keyword(s):  
Cloud Computing ◽  
Virtual Machine ◽  
Surveillance System ◽  
Video Stream ◽  
Visual Surveillance ◽  
Video Transcoding ◽  
Surveillance Systems ◽  
Surveillance Video ◽  
Private Cloud ◽  
And Storage

A large amount of surveillance videos and images need sufficient storage. In this article, an architecture of cloud-based surveillance systems and its modules will be designed, the Cloud-based Visual Surveillance System (CVSS) will be implemented on a private cloud using a Virtual Machine (VM). The users are able to link their cameras to the CVSS system so that the goal of this design can be achieved. The authors' CVSS system is able to push notification messages of captured videos to receivers, and their users could receive a surveillance video along with its events. The CVSS system fully makes use of the merits of cloud computing, which make it more advanced as stated in the evaluation section of this article. The contributions of this article are to be implemented in the CVSS system with: (1) video stream input, (2) intelligent visual surveillance, (3) real-time video transcoding and storage, (4) message pushing and media streaming output.

scholarly journals Generation and Annotation of Simulation-Real Ship Images for Convolutional Neural Networks Training and Testing

2021 ◽  
Vol 11 (13) ◽  
pp. 5931
Author(s):  
Ji’an You ◽  
Zhaozheng Hu ◽  
Chao Peng ◽  
Zhiqiang Wang
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Image Annotation ◽  
Three Dimensional ◽  
Image Data ◽  
Detection Algorithm ◽  
Simulation Software ◽  
High Quality ◽  
Annotation Method ◽  
Detection Of Objects

Large amounts of high-quality image data are the basis and premise of the high accuracy detection of objects in the field of convolutional neural networks (CNN). It is challenging to collect various high-quality ship image data based on the marine environment. A novel method based on CNN is proposed to generate a large number of high-quality ship images to address this. We obtained ship images with different perspectives and different sizes by adjusting the ships’ postures and sizes in three-dimensional (3D) simulation software, then 3D ship data were transformed into 2D ship image according to the principle of pinhole imaging. We selected specific experimental scenes as background images, and the target ships of the 2D ship images were superimposed onto the background images to generate “Simulation–Real” ship images (named SRS images hereafter). Additionally, an image annotation method based on SRS images was designed. Finally, the target detection algorithm based on CNN was used to train and test the generated SRS images. The proposed method is suitable for generating a large number of high-quality ship image samples and annotation data of corresponding ship images quickly to significantly improve the accuracy of ship detection. The annotation method proposed is superior to the annotation methods that label images with the image annotation software of Label-me and Label-img in terms of labeling the SRS images.

scholarly journals Is over-reporting a problem in mandatory communicable disease reporting? The taiwanese experience on Enterovirus 71

2014 ◽  
Vol 8 (10) ◽  
pp. 1294-1300
Author(s):  
Tseng Chu-Chun ◽  
Yang Che-Ming
Keyword(s):  
Communicable Disease ◽  
Enterovirus 71 ◽  
Surveillance Systems ◽  
Accuracy Rate ◽  
Cross Sectional ◽  
Systems Methodology ◽  
Disease Reporting ◽  
Enteroviral Infection ◽  
District Hospitals ◽  
The Government

Introduction: In Taiwan, severe enteroviral infections must be reported to the government within 24 hours to ensure that severe enterovirus 71 (EV71) infections can be detected early. The objective of this research was to ascertain whether over-reporting is a problem in mandatory disease-reporting systems. Methodology: A multiyear cross-sectional study methodology was applied based on secondary data analyses. Data from the national notifiable communicable disease surveillance system of Taiwan Centers for Disease Control were analyzed to assess the trends and factors influencing reporting accuracy. Results: From July 1999 to December 2008, 2,611 cases of severe enteroviral infection were reported in Taiwan. Among these cases, 1,516 were confirmed to be EV71 cases, and the remaining 1,095 were confirmed to be non-EV71 infections. The overall accuracy rate was 58%. The accuracy rate was 60%–70% higher during epidemics (2000–2002, 2005, and 2008) and high seasons than it was in other seasons. The accuracy rate was highest among medical centers and lowest among district hospitals. Conclusions: The results indicated that reports are more accurate during high seasons and peak years than during other periods. This might be attributable to the adequate level of specific educational programs for professionals when more cases occur, which could facilitate identification. Based on experiences in Taiwan, optimal training can ensure that surveillance systems are not inundated by false-positive reports.

Sign in / Sign up

Export Citation Format

Share Document