This study describes the development of a smart trash bin that separates and collects recyclables using a webcam and You Only Look Once (YOLO) real-time object detection in Raspberry Pi, to detect and classify these recyclables into their correct categories. The classification result rotates the trash bin lid and reveals the correct trash bin compartment for the user to throw away trash. The performance of the YOLO model was evaluated to measure its accuracy, which was 91% under an optimal computing environment and 75% when deployed in Raspberry Pi. Several Internet of Things hardware, such as ultrasonic sensors for measuring trash bin capacity and GPS for locating trash bin coordinates, are implemented to provide capacity monitoring controlled by Arduino Uno. The capacity and GPS information are uploaded to Firebase Database via theESP8266 Wi-Fi module. To deliver the capacity monitoring feature, the uploaded trash bin capacity information is displayed on the mobile application in the form of a bar level developed in the MIT App Inventor for the user to quickly take action if required. The system proposed in this study is intended to be implemented in a rural area, where it can potentially solve the recyclable waste separation problem.

This paper contains the development of a training service for foreigners to help them increase their ability to speak Korean. The service developed in this paper is implemented in the form of a mobile application that shows specific Korean sentences to the user for them to record themselves speaking the sentence. The objective is to generate the score automatically based on how similar the recorded voice with the actual sentence using Speech-To-Text (STT) engines and Sentence Transformers. The application is developed by selecting the four most commonly known STT engines with similar features, which are Google API, Microsoft Azure, Naver Clova, and IBM Watson, which are put into a Rest API along with the Sentence Transformer. The mobile application will record the user’s voice and send it to the Rest API. The STT engines will transcribe the file into a text and then feed it into a Sentence Transformer to generate the score based on their similarity. After measuring the response time and consistency as the performance evaluation by simulating a scenario using an Android emulator, Microsoft Azure with 1.13 s is found to be the fastest STT engine and Naver Clova is found to be the least consistent engine with nine different transcribe results.

Wanda Gusdya Purnama, Handoko Supeno, Anggoro Ari Nurcahyo, Ayi Purbasari, Aria Bisma Wahyutama, and Mintae Hwang. Journal of information and communication convergence engineering 2024;22:336-43. https://doi.org/10.56977/jicce.2024.22.4.336

Vehicle access to restricted and government-controlled sites requires verification of persistent vehicle and driver identity together with temporary authorization conditions. This study presents a blockchain-supported temporary credential scheme that separates persistent identity references from request-specific access conditions. A QR code is used as the presentation medium for a custom signed credential containing vehicle and driver identifiers, plate number, validity information, geofence parameters, and a nonce, while the blockchain maintains the corresponding credential hash, nonce, usage state, and revocation state. The scheme was evaluated through a bonded-zone case study using a public reference implementation and explicit trust assumptions. Six implemented conditions were tested: credential expiration, replay after successful use, out-of-geofence presentation, revocation, tampering of five signed payload fields, and cryptographic-key replacement while retaining the same prototype identifier. Across 55 recorded verification outcomes, all expired, replayed, out-of-geofence, revoked, and tampered credentials were rejected as expected. In the key-replacement experiment, all credentials generated with the previous key were rejected, whereas all newly generated credentials using the replacement key were accepted. Configuration-specific baseline experiments were also performed using an HS256 JWT implementation with centralized replay state and a blockchain implementation using ordinary vehicle and driver identifiers. The results demonstrate correct enforcement of the implemented credential-verification and blockchain-state rules under controlled prototype conditions, rather than comprehensive resistance to all real-world threats.

Aria Bisma Wahyutama and Mintae Hwang. Journal of information and communication convergence engineering 2026;24:32-40. https://doi.org/10.56977/jicce.2026.24.1.32