With the changes of the times and the rapid development of fast food, obesity has become a common concern in modern society. It is important for modern people to consider how to effectively manage their diet. Taking fast-food restaurants as an example, this study proposes a numerical label counting method and builds a food recognition system based on deep-learning-based YOLO object detection technology, in which 16 classes of well-known fast-food restaurants are trained, with a total of 1836 multi-label photos. When the detector accuracy rate is IoU = 0.6, mAP can reach 93.29%. In addition, the numerical-label-based multi-count method proposed in this study can quickly search and count a large number of food objects with a time complexity of O(Nlog2N). In a 50-search experiment of 75 items, the numerical-label-based multi-count method can be up to 2.152 times faster than the text label counting method. Moreover, this study uses the Eigen-CAM method of xAI technology to further examine the correctness of the results predicted by the detection model through visualization, where the Eigen-CAM method is one of class activation map (CAM) techniques for analyzing classification problems in computer vision. Finally, users can use the Android-based APP built by this study to identify and count food, quickly query the calories and nutritional content of food, and record information related to food intake, so as to provide users with automated diet management.

We present Dwell+, a method that boosts the effectiveness of typical dwell selection by augmenting the passive dwell duration with active haptic ticks which promptly drives rapid switches of modes forward through the user's skin sensations. In this way, Dwell+ enables multi-level dwell selection using rapid haptic ticks. To select a mode from a button, users dwell-touch the button until the mode of selection is haptically prompted. Our haptic stimulation design consists of a short 10ms vibrotacile feedback that indicates a mode arriving and a break that separates consecutive modes. We first tested the effectiveness of 170ms, 150ms, 130ms, and 110ms intervals between modes for a 10-level selection. The results reveal that 3-beats-per-chunk rhythm design, e.g., displaying longer 25ms vibrations initially for all three modes, could potentially achieve higher accuracy. The second study reveals significant improvement wherein a 94.5% accuracy was achieved for a 10-level Dwell+ selection using the 170ms interval with 3-beats-per-chunk design, and a 93.82% rate of accuracy using the more frequent 150ms interval with similar chunks for 5-level selection. The performance of conducting touch and receiving vibration from disparate hands was investigated for our final study to provide a wider range of usage. Our applications demonstrated implementing Dwell+ across interfaces, such as text input on a smartwatch, enhancing touch space for HMDs, boosting modalities of stylus-based tool selection, and extending the input vocabulary of physical interfaces.

Compared to the voice service over PSTN, people look forward to a competent Voice over IP (VoIP) service. The introduction of the SIP protocol has a profound impact on the VoIP world. Exploiting more SIP proxy servers can raise the service availability. However, how to assure the efficiency and availability of the service by such an improvement becomes a challenging issue. Based on above, we propose a probing-based name resolution scheme to achieve high availability and load balance for the SIP-based VoIP service. The idea is based on intercepting the prerequisite name resolution process in a typical client-server application in the IP network. We tag the probing mechanism onto the open source project Domain Name Relay Daemon (DNRD) to become a domain name resolution based load balancer (DN-LB). By our scheme, all SIP request messages from the SIP clients can be fairly distributed to a failure-proof SIP proxy server in the server farm without the necessity of using any additional costly intermediate network devices and changing the standard SIP architecture. Therefore, DN-LB can increase the service reliability and fault tolerance with a low cost for the SIP-based VoIP service.