This work describes the development of a vision-based tactile sensor system that utilizes the image-based information of the tactile sensor in conjunction with input loads at various motions to train the neural network for the estimation of tactile contact position, area, and force distribution. The current study also addresses pragmatic aspects, such as choice of the thickness and materials for the tactile fingertips and surface tendency, etc. The overall vision-based tactile sensor equipment interacts with an actuating motion controller, force gauge, and control PC (personal computer) with a LabVIEW software on it. The image acquisition was carried out using a compact stereo camera setup mounted inside the elastic body to observe and measure the amount of deformation by the motion and input load. The vision-based tactile sensor test bench was employed to collect the output contact position, angle, and force distribution caused by various randomly considered input loads for motion in X, Y, Z directions and $R_x{R}_y$ rotational motion. The retrieved image information, contact position, area, and force distribution from different input loads with specified 3D position and angle are utilized for deep learning. A convolutional neural network VGG-16 classification modelhas been modified to a regression network model and transfer learning was applied to suit the regression task of estimating contact position and force distribution. Several experiments were carried out using thick and thin sized tactile sensors with various shapes, such as circle, square, hexagon, for better validation of the predicted contact position, contact area, and force distribution.

The world’s elderly population continues to grow at an unprecedented rate, creating a need to monitor the safety of an aging population. One of the current problems is accurately classifying elderly physical activities, especially falling down, and delivering prompt assistance to someone in need. Owing to the advancements in deep learning research, vision based solutions are employed for action recognition. One such popular approach is human pose estimation based action recognition or fall detection. Nevertheless, due to a lack of large-scale elderly fall datasets and the continuation of numerous challenges such as varying camera angles, illumination, and occlusion accurately classifying falls has been a problematic. To address these problems, this research first carried out a comprehensive study of the AI Hub dataset collected from real lives of elderly people in order to benchmark the performance of state-of-the-art human pose estimation methods. Secondly, owing to the limited number of real datasets, augmentation with synthetic data was applied and performance improvement was validated based on changes in the degree of accuracy. Third, this study shows that a Transformer network applied to elderly action recognition outperforms LSTM-based networks by a noticeable margin. Lastly, by observing the quantitative and qualitative performances of different networks, this paper proposes an efficient solution for elderly activity recognition and fall detection in the context of surveillance cameras.

Developing a self-driving system is a challenging task that requires a high level of scene comprehension with real-time inference, and it is safety-critical. This study proposes a real-time memory efficient multitask learning-based model for joint object detection, drivable area segmentation, and lane detection tasks. To accomplish this research objective, the encoder-decoder architecture efficiently utilized to handle input frames through shared representation. Comprehensive experiments conducted on a challenging public Berkeley Deep Drive(BDD100 K) dataset. For further performance comparisons, a private dataset consisting of 30 K frames was collected and annotated for the three aforementioned tasks. Experimental results demonstrated the superiority of the proposed method's over existing baseline approaches in terms of computational efficiency, model power consumption and accuracy performance. The performance results for object detection, drivable area segmentation and lane detection tasks showed the highest 77.5 mAP50, 91.9 mIoU and 33.8 mIoU results on BDD100 K dataset respectively. In addition, the model achieved 112.29 fps processing speed improving both performance and inference speed results of existing multi-tasking models.