Robust Feature-Based Automated Multi-View Human Action Recognition System

ABSTRACT:

Automated human action recognition has the potential to play an important role in public security, for example, in relation to the multiview surveillance videos taken in public places, such as train stations or airports. This paper compares three practical, reliable, and generic systems for multiview video-based human action recognition, namely, the nearest neighbor classifier, Gaussian mixture model classifier, and the nearest mean classifier. To describe the different actions performed in different views, view-invariant features are proposed to address multiview action recognition. These features are obtained by extracting the holistic features from different temporal scales which are modeled as points of interest which represent the global spatial-temporal distribution. Experiments and cross-data testing are conducted on the KTH, WEIZMANN, and MuHAVi datasets. The system does not need to be retrained when scenarios are changed which means the trained database can be applied in a wide variety of environments, such as view angle or background changes. The experiment results show that the proposed approach outperforms the existing methods on the KTH and WEIZMANN datasets.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

• Willems et al. proposed the spatio-temporal domain which is an extension of the SURF descriptor.

• Schuldt et al. and Dollar et al. described sparse spatio-temporal features to deal with the complexity of human action recognition.

• Schuldt et al. proposed the representation of action using 3D spatio-temporal interest points captured from video frames. Schuldt also produced a histogram of informative words for each action adopting the codebook and bag-of-words (BOW) approach.

DISADVANTAGES OF EXISTING SYSTEM:

• Most of the early work assumes that the action is captured from a static viewpoint without any camera movement.

• These approaches only extend the system from a single viewpoint to a multi-view dataset.

• Existing approaches have difficulty ensuring the performance of the classifier when the viewpoint or environment changes.

PROPOSED SYSTEM:

The proposed system has following steps,

1. First, videos are acquiring from database.

2. Second, frames extraction is performing to further process.

3. Third, implementing of detection and localization of moving objects in each frame based on gaussian mixture model, prewitt filter and blob analysis.

4. Fourth, interest feature points are extracted from localized moving object and strongest key points are extracted based on Harris Spatio temporal corner detector.

5. Finally, Nearest Mean Classifier is proposed to recognize the human actions based on interest features.

ADVANTAGES OF PROPOSED SYSTEM:

• The rate of recognition accuracy is high.

• The computational time is less.

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

• System           : Pentium Dual Core.

• Hard Disk      : 120 GB.

• Monitor          : 15’’ LED

• Input Devices : Keyboard, Mouse

• Ram                : 1GB.

SOFTWARE REQUIREMENTS:

• Operating system : Windows 7.

• Coding Language : MATLAB.

• Tool                       : MATLAB R2013A

REFERENCE:

KUANG-PEN CHOU1, MUKESH PRASAD 2, DI WU2, NABIN SHARMA2, (Senior Member, IEEE), DONG-LIN LI3, YU-FENG LIN1, MICHAEL BLUMENSTEIN2, WEN-CHIEH LIN1, AND CHIN-TENG LIN2, (Fellow, IEEE), “Robust Feature-Based Automated Multi-View Human Action Recognition System”, IEEE Access 2018.