Two-Stage Feature Generator for Handwritten Digit Classification

Article Chemistry, Analytical

Air-GR: An Over-the-Air Handwritten Character Recognition System Based on Coordinate Correction YOLOv5 Algorithm and LGR-CNN

Yajun Zhang et al.

Summary: Traditional human-computer interaction technology based on input devices such as mice and keyboards cannot meet the advanced interaction needs of users. This paper proposes Air-GR, an over-the-air handwritten character recognition system, which combines the YOLOv5 algorithm and a lightweight convolutional neural network (LGR-CNN) for accurate and fast gesture recognition. Experimental results show that Air-GR can recognize any combination of 26 English letters and numbers with an accuracy of 95.24%.

SENSORS (2023)

Add to Collection

Article Chemistry, Analytical

Acoustic Sensing Based on Online Handwritten Signature Verification

Mengqi Chen et al.

Summary: In this paper, a handheld signature verification scheme called SilentSign is proposed. It leverages acoustic sensors in mobile devices to provide handy and safe paper-based signature verification services. By designing signal modulation strategies, developing a distance measurement algorithm, and training a verification model, SilentSign achieves efficient signature verification for users.

SENSORS (2022)

Add to Collection

Article Chemistry, Analytical

Deep-Learning-Based Character Recognition from Handwriting Motion Data Captured Using IMU and Force Sensors

Tsige Tadesse Alemayoh et al.

Summary: In this paper, a deep-learning-based smart digital pen that recognizes 36 alphanumeric characters was developed. Handwriting recognition was achieved using hand motion data captured by an inertial force sensor. The ViT network achieved a validation accuracy of 99.05% and showed promising performance in real-time validation.

SENSORS (2022)

Add to Collection

Article Chemistry, Analytical

The Hybrid Stylus: A Multi-Surface Active Stylus for Interacting with and Handwriting on Paper, Tabletop Display or Both

Cuauhtli Campos et al.

Summary: Paper continues to play an important role in the digital age, and combining it with digital medium can have significant benefits. However, digitizing handwritten information on paper is challenging due to the need for special paper, complex setups, and software. This paper presents a Hybrid Stylus that allows writing on physical paper while maintaining tablet interaction and digitizing handwritten information.

SENSORS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence