The advancements in computer vision-related technologies attract many researchers for surveillance applications, particularly involving the automated crowded scenes analysis such as crowd counting in a very congested scene. In crowd counting, the main goal is to count or estimate the number of people in a particular scene. Understanding overcrowded scenes in real-time is important for instant responsive actions. However, it is a very difficult task due to some of the key challenges including clutter background, occlusion, variations in human pose and scale, and limited surveillance training data, that are inadequately covered in the employed literature. To tackle these challenges, we introduce “SD-Net” an end-to-end CNN architecture, which produces real-time high quality density maps and effectively counts people in extremely overcrowded scenes. The proposed architecture consists of depthwise separable, standard, and dilated 2D convolutional layers. Depthwise separable and standard 2D convolutional layers are used to extract 2D features. Instead of using pooling layers, dilated 2D convolutional layers are employed that results in huge receptive fields and reduces the number of parameters. Our CNN architecture is evaluated using four publicly available crowd analysis datasets, demonstrating superiority over state-of-the-art in terms of accuracy and model size.

SD-Net: Understanding overcrowded scenes in real-time via an efficient dilated convolutional neural network

With the rapid development of the manufacturing industry, industrial automation equipment represented by computer numerical control (CNC) machine tools has put forward higher and higher requirements for the machining accuracy of parts. Compared with the multi-axis serial platform solution, the parallel platform solution is theoretically more suitable for high-precision machining equipment. There are many parallel platform solutions, but not one can provide a common physical platform to test the effectiveness of a variety of control algorithms. To achieve the goals, this paper is based on the Stewart six degrees of freedom parallel platform, and it mainly studies the platform construction. This study completed the mechanical structure design of the parallel platform. Based on the microprogrammed control unit (MCU) + pre-driver chip + three-phase full bridge solution, we have completed the circuit design of the motor driver. We wrote the program of MCU to drive six parallel robotic arms as well as the program of the parallel platform control center on the PC, and we completed the system joint debugging. The closed-loop control effect of the parallel platform workspace pose is realized.

/pdf/a-novel-architecture-of-a-six-degrees-of-freedom-parallel-319gcjcb.pdf

A Novel Architecture of a Six Degrees of Freedom Parallel Platform

Data is always a crucial issue of concern especially during its prediction and computation in digital revolution. This paper exactly helps in providing efficient learning mechanism for accurate predictability and reducing redundant data communication. It also discusses the Bayesian analysis that finds the conditional probability of at least two parametric based predictions for the data. The paper presents a method for improving the performance of Bayesian classification using the combination of Kalman Filter and K-means. The method is applied on a small dataset just for establishing the fact that the proposed algorithm can reduce the time for computing the clusters from data. The proposed Bayesian learning probabilistic model is used to check the statistical noise and other inaccuracies using unknown variables. This scenario is being implemented using efficient machine learning algorithm to perpetuate the Bayesian probabilistic approach. It also demonstrates the generative function for Kalman-filer based prediction model and its observations. This paper implements the algorithm using open source platform of Python and efficiently integrates all different modules to piece of code via Common Platform Enumeration (CPE) for Python. 

https://www.techscience.com/cmc/v72n1/46868/pdf

Prognostic Kalman Filter Based Bayesian Learning Model for Data Accuracy Prediction

Comprehensive analysis of energy efficient secure routing protocol over sensor network

Computer vision based fatigue detection using facial parameters

The Big data and cloud integration is a challenging Task. To enhance the data security issues, ABE can be deployed. In proposed model, a improved concept has been implemented and the integration of cloud and Big data is achieved. Security is the major threat for cloud computing applications. Every user has to feed user name, password, and primary key for Data access into the cloud data center. Data owner generates a new key to the users for accessing the data. Policy updating is also implemented in the proposed system, that is the accountability for the data access has also been implemented. In case of the change of policy, the altered data stored in the cloud is not affected. In addition to that, admin generates policy key based on the user’s profile. If any user tries to misbehave, an immediate alert is sent to the data owner. Data owner can change the policy key and access policy in the run time. Our system should be able to update its policy automatically.

A Novel Approach to View and Modify Data in Cloud Environment Using Attribute-Based Encryption

One of the primary tools for monitoring human movement and to prevent unwanted or unintended activities is surveillance camera. Nowadays, security management professional rely heavily on video surveillance to challenge crime and avert negative incidents that impact human society. Large numbers of surveillance camera installations have been increasing dramatically in the private and public sectors to monitor public activities. Video surveillance is one of the most effective methods to guarantee security. Fitting a surveillance camera simply transfers the captured video to security personnel. But the abnormal activities can be identified only by integrating an intelligent system to analyze the videos. Hence, this paper is motivated to design and implement an Intelligent Video Analytics Model (IVAM) also known as Human Object Detection (HOD) method for analyzing and detecting video-based abundant objects and abnormal human activities. IVAM can be deployed along with surveillance cameras in any public places like airport, hospital, shopping malls and railway station to automatically identify any abnormal event. IVAM is experimented with MATLAB software and the results are verified. The performance of IVAM is evaluated by comparing the obtained results with the existing approaches and it is proved that IVAM performs better compared to other contemporary methods in terms of accurately detecting the anomalies with less error rate and high classification accuracy.

An intelligent video analytics model for abnormal event detection in online surveillance video

https://www.techscience.com/iasc/v32n2/45602/pdf

A. Balasundaram

Papers

A Novel Approach to View and Modify Data in Cloud Environment Using Attribute-Based Encryption

An intelligent video analytics model for abnormal event detection in online surveillance video

Prognostic Kalman Filter Based Bayesian Learning Model for Data Accuracy Prediction

Abnormality Identification in Video Surveillance System using DCT

Computer vision based fatigue detection using facial parameters