Human action recognition using Meta-Cognitive Neuro-Fuzzy Inference System

Abstract

In this paper, we propose a Meta-Cognitive NeuroFuzzy Inference System (McFIS) for accurate detection of human actions from video sequences. In this paper, we employ optical flow based features as they can represent information from local pixel level to global object level between two consecutive image planes. The functional relationship between these optical flow based features and action classes is approximated using McFIS classifier. The sequential learning algorithm is developed based on the principles of self-regulation observed in human metacognition. McFIS decides on what-to-learn, when-to-learn and how-to-learn based on the knowledge stored in the classifier and the information contained in the new training sample. The sequential learning algorithm of McFIS is controlled and monitored by the meta-cognitive components which uses class-specific and knowledge based criteria along with self-regulatory thresholds to decide on one of the following strategies: a) sample deletion b) sample learning and c) sample reserve. Performance of proposed McFIS based human action recognition system is evaluated using benchmark Weizmann and KTH video sequences. The simulation results are compared with well known support vector machine classifier and also with state-of-the-art action recognition results reported in the literature. The results clearly indicates McFIS action recognition system achieves better performances with minimal computational effort.

A Survey on Study of Various Machine Learning Methods for Classification

“…[1] Controlling of aircraft using neurons also applied using MRAN. [2] Besides this it also applied for signal processing [4].…”

“…It has been applied in various fields such as developing a simulator for medical diagnostics, image processing etc. [16][17]19]. The above Table 1 summarizes the merits and demerits of the machine learning algorithms taken for discussion.…”

Fuzzy‐Valued Intensity Measures for Near‐Fault Pulse‐Like Ground Motions

“…The definition method is also the focus of research (Reuter, 2011;Yan and Ma, 2012b) although in most cases the membership function can be defined empirically. More recently, neural networks and fuzzy logic has been integrated to solve engineering problems (Samant and Adeli, 2001;Theodoridis et al, 2012;Subramanian and Suresh, 2012;Liu and Er, 2012), and fuzzy theory is used in data mining (Jiang and Adeli, 2003;Ma et al, 2011).…”

A Metacognitive Fully Complex Valued Functional Link Network for solving real valued classification problems

“…Also, meta-cognitive sequential learning for interval type 2 neuro fuzzy inference system has been developed in [40][41][42]. Moreover, human action recognition [43] and human emotion recognition [44] classification problems are solved using meta-cognitive neuro fuzzy inference system. In order to overcome the drawback existing with the real valued meta-cognitive interval type 2 neuro fuzzy inference system, a complex valued meta-cognitive interval type 2 neuro fuzzy inference system has been proposed in [45].…”