Sunder Ali Khowaja

Hierarchical classification method based on selective learning of slacked hierarchy for activity recognition systems

Abstract Physical activity recognition using wearable sensors has gained significant interest from researchers working in the field of ambient intelligence and human behavior analysis. The problem of multi-class classification is an important issue in the applications which naturally has more than two classes. A well-known strategy to convert a multi-class classification problem into binary sub-problems is the error-correcting output coding (ECOC) method. Since existing methods use a single classifier with ECOC without considering the dependency among multiple classifiers, it often fails to generalize the performance and parameters in a real-life application, where different numbers of devices, sensors and sampling rates are used. To address this problem, we propose a unique hierarchical classification model based on the combination of two base binary classifiers using selective learning of slacked hierarchy and integrating the training of binary classifiers into a unified objective function. Our method maps the multi-class classification problem to multi-level classification. A multi-tier voting scheme has been introduced to provide a final classification label at each level of the solicited model. The proposed method is evaluated on two publicly available datasets and compared with independent base classifiers. Furthermore, it has also been tested on real-life sensor readings for 3 different subjects to recognize four activities i.e. Walking, Standing, Jogging and Sitting. The presented method uses same hierarchical levels and parameters to achieve better performance on all three datasets having different number of devices, sensors and sampling rates. The average accuracies on publicly available dataset and real-life sensor readings were recorded to be 95% and 85%, respectively. The experimental results validate the effectiveness and generality of the proposed method in terms of performance and parameters.

Facial expression recognition using two-tier classification and its application to smart home automation system

With the convergence of smart technologies and advancement in electronic equipment the concept of smart home system swiftly escalates. The idea is to automate the home appliances according to the user requirements without human intervention. After a long tiring day and heavy workloads user will not be in a state of taking out its mobile phone and pressing the buttons for controlling home appliances. Several methods have been proposed in the design of such systems using sensors, biometrics and face detection. This paper proposes a method for detecting human emotions by taking into account the complete facial analysis, suggesting that the emotions can accurately be determined by analyzing eyes, nose and lips separately hence covering a wide range of emotions. The classification is carried out by acquiring the image of user followed by the face detection and segmenting the region of interests (ROI) i.e. eyes, nose and lips for further analysis of emotions. Principle Component Analysis (PCA) along with feature extraction techniques and Support Vector Machines (SVMs) are used for classification of emotion for the said automation system. Policies have been implemented in Java to simulate the home automation environment for testing and validation. At the instant this system has been tested on a single user with 4 basic emotions i.e. sad, anger, happiness and neutral, but this study can be a basis to develop an automated system with variety of emotions for multiple users.

Supervised method for blood vessel segmentation from coronary angiogram images using 7-D feature vector

With the recent advancement in medical image processing field and sophisticated simulation tools it has been possible to acquire useful information from raw images for different parts of the body. Coronary artery segmentation is the fundamental component which extract significant features from angiogram images. Cardiac catheterization is an invasive diagnostic procedure that provides important information about the structure and function of heart. The procedure usually involves X-ray images of heart, arteries using coronary angiography. The resultant images (coronary angiogram) are considered as best of way to diagnose cardiac heart disease. The main focus of coronary angiography is to find the blockage in major blood vessels, however if the blockage is not found in large blood vessels and patient persists to have pain (angina) then it is concluded that the patient is having micro vascular disease (MVD). MVD is caused by blockage or narrowing of small blood vessels in heart, unfortunately there is no specific test to diagnose MVD but it is common in people having diabetes and blood pressure. This paper proposes an automated method of vessel segmentation from coronary angiogram images using radial basis function and moment invariant-based features to extract the small blood vessel for diagnosis of MVD. Experimental results show that the proposed method is capable of extracting small blood vessels from coronary artery and can be a basis to identify key characteristics for MVD. The dataset of angiogram images have been provided by ISRA University Hospital and MATLAB is used for implementing the proposed method.

A Novel Curvature Feature Embedded Level Set Method for Image Segmentation of Coronary Angiograms

Segmentation methods in medical image processing are usually distorted by low contrast and intensity inhomogeneity. There are several image segmentation methods which are based on region based segmentation. But these algorithms mostly depend on the quality of the image. This paper gives an improved level set method for image segmentation to reduce the effect of noise. In order to achieve this, curvature feature energy function in standard level set energy function has been used. The proposed method is being applied on heart angiograms provided by Cardiac Department ISRA University Hospital, Pakistan. Extensive evaluation of these images depicts the robustness and efficiency of the proposed method over the previous work. Moreover, this method gives better trade-off between accuracy and implementation time over the related work.

A framework for retinal vessel segmentation from fundus images using hybrid feature set and hierarchical classification

Retinal blood vessels play an imperative role in detection of many ailments, such as cardiovascular diseases, hypertension, and diabetic retinopathy. The automated way of segmenting vessels from retinal images can help in early detection of many diseases. In this paper, we propose a framework based on hybrid feature set and hierarchical classification approach to segment blood vessels from digital retinal images. Firstly, we apply bidirectional histogram equalization on the inverted green channel to enhance the fundus image. Six discriminative feature extraction methods have been employed comprising of local intensities, local binary patterns, histogram of gradients, divergence of vector field, high-order local autocorrelations, and morphological transformation. The selection of feature sets has been carried out by classifying vessel and background pixels using random forests and evaluating the segmentation performance for each category of features. The selected feature sets are then used in conjunction with our proposed hierarchical classification approach to segment the vessels. The proposed framework has been tested on the DRIVE, STARE, and CHASEDB1 which are the benchmark datasets for retinal vessel segmentation methods. The results obtained from the experimental analysis show that the proposed framework can achieve better results than most state-of-the-art methods.

Contextual activity based Healthcare Internet of Things, Services, and People (HIoTSP): An architectural framework for healthcare monitoring using wearable sensors

Abstract Healthcare industry is gaining a lot of attention due to its technological advancement and the miniaturization in the form of wearable sensors. IoT-driven healthcare industry has mainly focused on the integration of sensors rather than the integration of services and people. Nonetheless, the framework for IoT-driven healthcare applications are significantly lacking. In addition, the use of semantics for ontological reasoning and the integration of mobile applications into a single framework have also been ignored in many existing studies. This work presents the implementation of Healthcare Internet of Things, Services, and People (HIoTSP) framework using wearable sensor technology. It is designed to achieve the low-cost (consumer devices), the easiness to use (interface), and the pervasiveness (wearable sensors) for healthcare monitoring along with the integration of services and agents like doctors or caregivers. The proposed framework provides the functionalities for data acquisition from wearable sensors, contextual activity recognition, automatic selection of services and applications, user interface, and value-added services such as alert generation, recommendations, and visualization. We used the publicly available dataset, PAMAP2 which is a physical activity monitoring dataset, for deriving the contextual activity. Fall and stress detection services are implemented as case studies for validating the realization of the proposed framework. Experimental analysis shows that we achieve, 87.16% accuracy for low-level contextual activities and 84.06%–86.36% for high-level contextual activities, respectively. We also achieved 91.68% and 82.93% accuracies for fall and stress detection services, respectively. The result is quite satisfactory, considering that all these services have been implemented using pervasive devices with the low-sampling rate. The real-time applicability of the proposed framework is validated by performing the response time analysis for both the services. We also provide suggestions to cope with the scalability and security issues using the HIoTSP framework and we intend to implement those suggestions in our future work.

Energy efficient mobile user tracking system with node activation mechanism using wireless sensor networks

User localization in wireless sensor networks have been given a great attention in recent times and also considered to be one of the promising applications. Many approaches for the same have been proposed based on range based and range free mechanisms for localizing the user. Similarly, tracking the user in a sensor field has also been given equal importance in association with localization. Techniques employing both these methods are mostly based on static anchor nodes or scheduling system which compromises on the lifetime of wireless sensor network. Considering the constraint of network lifetime this paper proposes localizing and tracking method with an activation scheme for tracking the mobile node efficiently while increasing the network lifetime. The system has been tested on two scenarios suggesting that the proposed method can provide the flexibility to the system which could be adjusted with reference to the user requirements. The first scenario suggests that all the nodes are activated only when the mobile user enters but as the user is localized all the nodes will get deactivated except the concerned nodes. The second scenario suggests that a predefined deployment strategy is provided with only 10% of activated nodes. The experimental results show that the proposed system achieves a better trade-off in terms of accuracy and computational complexity for single mobile node tracking.