Recognition By Using Adaptive Neural Fuzzy Computer Science Essay

Published Date: 02 Nov 2017

Abstract

View invariant action recognition method based on adaptive neural fuzzy inference system (ANFIS) is proposed. ANFIS is an intelligence method which aggregates both fuzzy stage and neural networks. It will determine the parameter values automatically according to the data. Human body posture prototypes are identified by self organizing map. Fuzzy inference system is proposed to identify the action classification. This method maps, set of input data onto a set of desired output and it matches the data from the training and testing phase. Bayesian framework to recognize the various kinds of activities and produes the recognition results.The proposed method is able to discover continuous instances of similar action performed by several people in various view points.

Keywords-ANFIS, bayesian frameworks, human action recognition, multilayer perceptrons, self organizing map, view invariance.

Introduction

Human movement recognition is the most widely used approach for video analysis. It is considered as an important problem for many applications such as visual surveillance [2], human-machine interaction [3], analysis of video events [4], entertainment and sports etc. The term action refers to a single period of human motion pattern for a period of time. Action is discriminate from activity. An activity is continuous event of small atomic actions. For example the activity jogging consists of the following actions walk, run, jump etc. Recognizing the human action is a very challenging problem because the actions can look very different depending upon the context such as similar actions with various garbs, action may be performed by different kinds of people in multiple viewpoints or different people performed the same action but it may appears in various ways [1].

Representation of human action is used to match the similarity of all human body poses by a self organizing map (SOM) in a neural network. In the training phase SOM is used to train the data in the posture images and represents the actions also. In the testing phase Adaptive neural fuzzy inference system is used to test the data in the posture images and produces the aggregation results for human actions. It utilizes the fuzzy rules and the membership functions parameters [5]. For action classification Fuzzy inference system (FIS) is proposed. It automatically calculates the parameter values without human direct interference. This method is very efficient to reduce the computational effect. Bayesian Framework is to recognize the unknown actions and also produces the combined recognition results with high classification accuracy [6].

RELATED WORK

S. Ali and M. Shah proposed kinematic features for Action recognition. It represents the complex human action in videos. Kinematic features not view invariant because the same action viewed by different viewing angle. Occlusion will also affect the performance of the action [7]. H. J. Seo and P. Milan far proposed regression kernel analysis. It captures the data even in the presence of misrepresentation of action and error present in the data. It also finds the similarity actions and does not need prior knowledge about actions [4]. M. Ahmad and S. Lee proposed Hidden Markov Model. It recognizes the actions from random view instead of any particular view. It is used to represent the actions from various viewing angles.HMM for movement recognition is used to create the time series data. It is the most widely used approach for speech and word recognition [8].

N.Gkalelis et al proposed fuzzy vector quantization (FVQ) and linear discriminant analysis (LDA). These methods have the capability to distinguish the similar movements. LDA reduces the dimensionality of the multiview movement video features. This method is powerful because low dimensionality features produce the recognition accuracy. It finds only the linear combination of features in a classes of objects or events .FVQ is used to match the input image vector to basic movement pattern space. It also increases the quality of the input vector [9]. F. Lv and R. Nevatia proposed Pyramid Match Kernel algorithm. It provides the comparable rate between two same characteristics of images.It achieves comparable result and lower computational cost. It also reduces the difficulty of movement recognition problem. But the single view action classification needs large number of parameter to solve the ambiguity of the classification [10].

S. Yu et al proposed appearance based gait recognition.It is valuable for robust gait recognition system.This method is not suitable to recognize the human action from side view and also from various viewing angles [11].D. Weinland et al proposed principal component analysis (PCA).It is mostly used to decrease the high dimensional image features into low dimensional image features. It is useful for view invariant recognition for larger class of primitive actions.It does not perform linear separation and linear regression of classes and it does not perform the similar human actions also [8].

.

MATERIALS AND METHODS

A. Experimental Setup

Human action recognition is automated detection of ongoing events from video data. Action recognition is the finding of video segments containing such actions. Video segment is used to display the properties of the actions. The video sequences are collected from Weizmann datasets [12].The video sequences are converted into frames and stored in the database. It contains the actions such as Bend, walk, run, jump in place and wave two hands etc.

The proposed method consists of identification of posture prototypes, testing of data with ANFIS method, Action classification and action recognition. The overview of the proposed method is as shown in the fig.1.

The diagram represents SOM is used to train the data in the training phase. After classification of SOM, Fuzzy inference system is used to test the data. Finally Bayesian framework is to recognize the actions.

Fig 1.Overview of the proposed method

B.Methods

Preprocessing phase

In action recognition, elementary action video sequences are converted into video frames. Moving object segmentation techniques [13, 14] are used to create binary images. Background elimination is a widely used approach for detecting the moving object. After the background elimination, person’s body is extracted and produces the binary posture frames with the same size. Binary image frames of five movements bend, walk, run, jump in place and wave two hands by using the edge detection method in Fig 2. Continuous movement of bend postures is shown in Fig 3.

C:\Documents and Settings\acer\Desktop\Questions\anitha output\bend\bend10 correct.JPGC:\Documents and Settings\acer\Desktop\Questions\anitha output\walk\walk4.JPGC:\Documents and Settings\acer\Desktop\Questions\anitha output\run\run4.JPGC:\Users\keerthana\Desktop\jump in place.JPGK:\wave2.JPG

Fig 2.Binary image frames of five actions

C:\Documents and Settings\acer\Desktop\anitha output\bend\10bend.JPG

Fig 3.Continuous movement of bend postures

Identification of posture prototypes

In the training phase videos sequences of the posture images are clustered to a fixed number of classes using a self organizing map (SOM) algorithm [15]. It is a special class of neural networks. It uses the unsupervised learning method which does not need any external resources for getting the desired output. The SOM is used to identify and grouping different portion of images with similar features. An output neuron with smallest value is determined as the winner in competition that unit is called best matching unit.

The basic process for training the data based on SOM has the following steps.

1) Intialization: Weights are initialized randomly.

2) Sampling: Produce the sample X and give it to the network.

3) Similarity Matching: The winning neuron N is mapped with the weight of the input vector. It is considered as the best matching neuron.

N = argmin (j) {X-Wj}

4) Updating: Adjust the parameter of the neighbourhood function.

∆W = γ. hij (X - Wj)

Where hij is the neighbourhood function, γ is the learning rate dependent on time.

The algorithm is trained up to 100 iterations. This procedure is applied multiple times for training the data’s which were not trained. Finally it represents the actions.

Testing and classification of data with ANFIS method

In this phase, the user gives an input posture image for which the corresponding output image is tested. Here the input data is normalized and then checked with the ANFIS method. It uses the sugeno type fuzzy inference system for training routine. It utilizes the automatic identification of fuzzy rules and membership function parameters [5].

FIS classifier

Fuzzy Inference System using a strategy of hybrid approach of adaptive neural-fuzzy inference system. It is the hybrid approach to identify the parameter of sugeno type fuzzy inference system. It is the complicated method but it gives the probable results which are more efficient. In action classification, FIS classifier completed the training of data upto 100 epochs. Once trained, FIS is used to classify the each testing data in the posture images and classify the actions depending upon the images are already trained by SOM.Finally it produces the most repeated occurrence of actions.

Action Recognition

In the action recognition phase, video frames are segmented by using the background elimination method and the features are also extracted. The input frame is compared with the posture retained in the database. If a same posture is obtained, the posture is allocated to the label name of the current frame. Otherwise the new label name is assigned to the current frame of the posture which is retained in the database.

In the Bayesian framework case, the human actions are fed to the FIS classifier to recognize the corresponding action that computes the most combined recognition rate depending on the Bayesian approach [6].It produces the combined recognition results with high classification accuracy. The most probable recognition rate for confusion matrix is represented in Table 1.Finally it can recognize the action such as bend, walk and run etc. A recognition rate obtained for Bayesian approach is 86.66%.

RESULT AND ANALYSIS

The results and discussions of the human action recognition is based on Bayesian approach. There are two phases in the recognition method. In the training phase, SOM is trained and matches the similarity of all human actions. In the testing phase, FIS is used to test the data in the posture images and produces the aggregation results for human actions.

In action recognition video sequences are collected from the Weizmann datasets. Here 20 videos from the Weizmann datasets are used for action recognition. Each video describes one human performing one action. The video sequence are converted into frames and stored in the database. It contains the action such as bend, walk and run, jump in place and wave two hands etc.

The input image is taken from the database as shown in Fig 3(a).The grayscale image is converted into a binary image using edge detection method. It detects the wide range of edges in image. The binary image is as shown in Fig 3(b).The binary image is segmented for clearly represent the action. By using segmentation techniques actions are easier to analyze. It is also used for extracting foreground from background model. The segmented image is as shown in fig 3(c).The input image is matched with the actions in the database. Here the input image is matched with the posture retained in the database. If a same posture is obtained, the posture is allocated to the label name of the current frame. Otherwise the new label name is assigned to the current frame of the posture which is retained in the database. Finally matches the similarity of the action and recognize the actions such as bend, walk and run.

C:\Documents and Settings\acer\Desktop\Questions\anitha output\bend\bend12.JPGC:\Documents and Settings\acer\Desktop\Questions\anitha output\bend\bend10 correct.JPGC:\Documents and Settings\acer\Desktop\Questions\anitha output\bend\bend11.JPGC:\Documents and Settings\acer\Desktop\Questions\anitha output\bend\bend12.JPG

C:\Documents and Settings\acer\Desktop\Questions\anitha output\walk\walk2.JPG C:\Documents and Settings\acer\Desktop\Questions\anitha output\walk\walk4.JPGC:\Documents and Settings\acer\Desktop\Questions\anitha output\walk\walk1.JPGC:\Documents and Settings\acer\Desktop\Questions\anitha output\walk\walk2.JPG

C:\Documents and Settings\acer\Desktop\Questions\anitha output\run\run1.JPG C:\Documents and Settings\acer\Desktop\Questions\anitha output\run\run4.JPG C:\Documents and Settings\acer\Desktop\Questions\anitha output\run\run2.JPGC:\Documents and Settings\acer\Desktop\Questions\anitha output\run\run1.JPG

(b) (c) (d)

Fig 3. Action recognition (a) Input image (b) Binary image

(c) Segmented image (d) Matched image

Analysis

Bayesian approach is used to recognize the action and the result is presented in table 1 by using the confusion matrix [16]. It consists of information about known class and predicted class. In this matrix rows represent the known defined classes and columns of the matrix represent the predicted classes. The diagonal values are classified perfectly and the off-diagonal values are incorrectly classified.

TABLE I

Confusion matrix for three actions

Bend

Walk

Run

Bend

19

1

0

Walk

2

16

2

Run

1

2

17

Overall correct classification rate is equal to 86.66% for Bayesian approach. Action which contains different body poses like bend is almost perfectly classified.Similar body poses like walk and run are difficult to be correctly classified.

Performance Metrics

Performance metrics compare the strength and weakness of different classifiers by computing the precision, recall and F1 metric [16].Performance metrics and accuracy results are described in the following.

Accuracy

It is the measure of total number of predictions that were perfectly classified.

Accuracy = (TP+TN) / ( TP+ TN + Fp +FN)

The overall accuracy of the human action recognition is 86.66% as shown in table II.

Precision

It is the measure of specific cases predicted based on positive classes.

Precision = TP / (TP+FP)

Recall

It is the measure of positive cases that were correctly calculated. It is also called sensitivity. It is similar to the true positive rate.

Recall = TP / (TP+FN)

TABLE II

Performance Metrics

Metrics

Bend

Walk

Run

Precision

Recall

F1

Similarity

Specificity

0.8636

0.9500

0.9047

0.8260

0.9250

0.8421

0.8000

0.8205

0.6956

0.9250

0.8947

0.8500

0.8718

0.7727

0.9500

F1 metric

Figure of metric or F measure is the weighted mean of precision and recall.

F1 metric = 2(Recall*Precision) / (Recall + Precision)

Similarity

It is the measure of calculates the two or more different actions from the database.

Similarity = TP / (Tp+FN +FP)

Specificity

It is the measure of negative cases classified correctly. It is same as the true negative rate.

Specificity = TN / (TN+FP)

Performance metrics for human action recognition is as shown in table II.

V.CONCLUSION AND FUTURE WORK

View invariant action recognition method based on adaptive neural fuzzy inference system to solve the generic action recognition problem. ANFIS is the very useful tools to train the images. It is a quick and straightforward way of input selection. SOM is constructed from the dataset processing and training the data and the input query is tested which is based on ANFIS. FIS classifier is used for classifying the given actions. It measures the similarity between images and produces the classification results. Bayesian approach is to recognize the human actions using a single video samples. This method also recognizes the continuous human action.

In future, this method can detect the human interaction between persons and also calculate the abnormal representation of human actions.