This paper presents research focusing on visualization and pattern recognition based on computer science. Although deep neural networks demonstrate satisfactory performance regarding image and voice recognition, as well as pattern analysis and intrusion detection, they exhibit inferior performance towards adversarial examples. Noise introduction, to some degree, to the original data could lead adversarial examples to be misclassified by deep neural networks, even though they can still be deemed as normal by humans. In this paper, a robust diversity adversarial training method against adversarial attacks was demonstrated. In this approach, the target model is more robust to unknown adversarial examples, as it trains various adversarial samples. During the experiment, Tensorflow was employed as our deep learning framework, while MNIST and Fashion-MNIST were used as experimental datasets. Results revealed that the diversity training method has lowered the attack success rate by an average of 27.2 and 24.3% for various adversarial examples, while maintaining the 98.7 and 91.5% accuracy rates regarding the original data of MNIST and Fashion-MNIST.
The growing use of collaborative robots (cobots) in industrial environments raises concerns about worker safety; flexible and adaptive systems are required to predict and prevent real-time cobot-worker collisions. This study proposes a digital-twin-based, cobot-worker collision risk-prediction system that utilizes the Unity game-development engine, robot operating system (ROS), and ultra-wideband (UWB) communication. The proposed system incorporates three modules: 1) real-time cobot-worker data capture, 2) digital-twin-based dynamic collision-zone management, and 3) collision-risk prediction. The core of the system is a data-driven cobot-worker digital twin that reflects the movements of the cobot and worker using ROS and UWB communication. This dynamic approach addresses the limitations of fixed safety-zone configurations by considering the cobot trajectory and the worker path projection in real time within the digital-twin modules. The system demonstrated low latency in data transmission from ROS to Unity (average 13 ms) and from UWB to Unity (average 140 ms), through implementation and evaluation. It also accurately predicted the minimum distance between the worker and cobot, accounting for varying worker speeds and future prediction times. This research offers a promising solution for enhancing worker safety in collaborative robotic environments by providing a real-time, digital-twin-based collision risk prediction system with dynamic collision zone management.
This paper proposes a pseudo-haptic interface that depicts the virtual weights of dumbbells in a virtual gymnasium. When a user performs a dumbbell biceps curl, he/she fixes the elbow joint as a standard joint and lifts the dumbbell, with its movement trajectory represented as a circular arc. The trajectories and velocity of dumbbell bicep curls differ depending on human physiological characteristics. Therefore, the proposed system provides an adaptable exercise area and force visualization of virtual dumbbells using a velocity-based pseudo-haptic interface and computer vision-based tracking method. The system recognizes the position and rotation of joints related to a dumbbell biceps curl with the implementation of density-based spatial clustering of applications with noise (a clustering algorithm) and resizes the radius and angle of an integrated force circular gauge. Furthermore, when a user lifts a dumbbell, the system recognizes, using linear regression, the current position and lifting force of the virtual dumbbell and visualizes the current lifting force with a guided movement trajectory to match the lifting force. Experimental results show that the proposed pseudo-haptic interface increased weight perception and usability by up to 30% compared to conventional methods (p < 0.05).
1 November 2022 in IEICE Transactions on Information and Systems
Deep neural networks (DNNs) provide excellent services in machine learning tasks such as image recognition, speech recognition, pattern recognition, and intrusion detection. However, an adversarial example created by adding a little noise to the original data can result in misclassification by the DNN and the human eye cannot tell the difference from the original data. For example, if an attacker creates a modified right-turn traffic sign that is incorrectly categorized by a DNN, an autonomous vehicle with the DNN will incorrectly classify the modified right-turn traffic sign as a U-Turn sign, while a human will correctly classify that changed sign as right turn sign. Such an adversarial example is a serious threat to a DNN. Recently, an adversarial example with multiple targets was introduced that causes misclassification by multiple models within each target class using a single modified image. However, it has the weakness that as the number of target models increases, the overall attack success rate decreases. Therefore, if there are multiple models that the attacker wishes to attack, the attacker must control the attack success rate for each model by considering the attack priority for each model. In this paper, we propose a priority adversarial example that considers the attack priority for each model in cases targeting multiple models. The proposed method controls the attack success rate for each model by adjusting the weight of the attack function in the generation process while maintaining minimal distortion. We used MNIST and CIFAR10 as data sets and Tensorflow as machine learning library. Experimental results show that the proposed method can control the attack success rate for each model by considering each model's attack priority while maintaining minimal distortion (average 3.95 and 2.45 with MNIST for targeted and untargeted attacks, respectively, and average 51.95 and 44.45 with CIFAR10 for targeted and untargeted attacks, respectively).
1 March 2020 in Journal of the Korea Computer Graphics Society
사단법인 한국컴퓨터그래픽스학회의 논문지로서 그래픽스 분야와 그에 연관된 연구영역의 새로운 발견과 최첨단 연구결과를 발표하는 토론의 장입니다. 학회를 대표하는 학술지인 논문지는 1995년에 논문지 발간을 시작으로 2000년부터 3월, 6월, 9월, 12월 총 4회 발간되었고, 2015년부터는 KCGS 학술대회 특별호를 7월에 발간하면서 총 5회 발간되고 있습니다.