Latest advancements in vision technology offer an evident impact on multi-object recognition and scene understanding. Such sceneunderstanding task is a demanding part of several technologies, like augmented reality-based scene integration, robotic navigation, autonomous driving, and tourist guide. Incorporating visual information in contextually unified segments, convolution neural networks-based approaches will significantly mitigate the clutter, which is usual in classical frameworks during scene understanding. In this paper, we propose a convolutional neural network (CNN) based segmentation method for the recognition of multiple objects in an image. Initially, after acquisition and preprocessing, the image is segmented by using CNN. Then, CNN features are extracted from these segmented objects, and discrete cosine transform (DCT) and discrete wavelet transform (DWT) features are computed. After the extraction of CNN features and computation of classical machine learning features, fusion is performed using a fusion technique. Then, to select the minimal set of features, genetic algorithm-based feature selection is used. In order to recognize and understand the multi-objects in the scene, a neuro-fuzzy approach is applied. Once objects in the scene are recognized, the relationship between these objects is examined by employing the object-to-object relation approach. Finally, a decision tree is incorporated to assign the relevant labels to the scenes based on recognized objects in the image. The experimental results over complex scene datasets including SUN Red Green Blue-Depth (RGB-D) and Cityscapes' demonstrated a remarkable performance.

With the dramatic increase in video surveillance applications and public safety measures, the need for an accurate and effective system for abnormal/suspicious activity classification also increases. Although it has multiple applications, the problem is very challenging. In this paper, a novel approach for detecting normal/abnormal activity has been proposed. We used the Gaussian Mixture Model (GMM) and Kalman filter to detect and track the objects, respectively. After that, we performed shadow removal to segment an object and its shadow. After object segmentation we performed occlusion detection method to detect occlusion between multiple human silhouettes and we implemented a novel method for region shrinking to isolate occluded humans. Fuzzy c-mean is utilized to verify human silhouettes and motion based features including velocity and optical flow are extracted for each identified silhouettes. Gray Wolf Optimizer (GWO) is used to optimize feature set followed by abnormal event classification that is performed using the XG-Boost classifier. This system is applicable in any surveillance application used for event detection or anomaly detection. Performance of proposed system is evaluated using University of Minnesota (UMN) dataset and UBI (University of Beira Interior)-Fight dataset, each having different type of anomaly. The mean accuracy for the UMN and UBI-Fight datasets is 90.14% and 76.9% respectively. These results are more accurate as compared to other existing methods.

No abstract

Virtual reality is an emerging field in the whole world. The problem faced by people today is that they are more indulged in indoor technology rather than outdoor activities. Hence, the proposed system introduces a fitness solution connecting virtual reality with a gaming interface so that an individual can play first-person games. The system proposed in this paper is an efficient and cost-effective solution that can entertain people along with playing outdoor games such as badminton and cricket while sitting in the room. To track the human movement, sensors Micro Processor Unit (MPU6050) are used that are connected with Bluetooth modules and Arduino responsible for sending the sensor data to the game. Further, the sensor data is sent to a machine learning model, which detects the game played by the user. The detected game will be operated on human gestures. A publicly available dataset named IM-Sporting Behaviors is initially used, which utilizes triaxial accelerometers attached to the subject's wrist, knee, and below neck regions to capture important aspects of human motion. The main objective is that the person is enjoying while playing the game and simultaneously is engaged in some kind of sporting activity. The proposed system uses artificial neural networks classifier giving an accuracy of 88.9%. The proposed system should apply to many systems such as construction, education, offices and the educational sector. Extensive experimentation proved the validity of the proposed system.

Independent human living systems require smart, intelligent, and sustainable online monitoring so that an individual can be assisted timely. Apart from ambient assisted living, the task of monitoring human activities plays an important role in different fields including virtual reality, surveillance security, and human interaction with robots. Such systems have been developed in the past with the use of various wearable inertial sensors and depth cameras to capture the human actions. In this paper, we propose multiple methods such as random occupancy pattern, spatio temporal cloud, waypoint trajectory, Hilbert transform, Walsh Hadamard transform and bone pair descriptors to extract optimal features corresponding to different human actions. These features sets are then normalized using min-max normalization and optimized using the Fuzzy optimization method. Finally, the Masi entropy classifier is applied for action recognition and classification. Experiments have been performed on three challenging datasets, namely, UTD-MHAD, 50 Salad, and CMU-MMAC. During experimental evaluation, the proposed novel approach of recognizing human actions has achieved an accuracy rate of 90.1% with UTD-MHAD dataset, 90.6% with 50 Salad dataset, and 89.5% with CMU-MMAC dataset. Hence experimental results validated the proposed system.