Results 51 to 60 of about 427 (175)

Data‐Driven Materials Science for Energy‐Sustainable Applications

Advanced Materials, EarlyView.
Data‐driven approaches powered by artificial intelligence are transforming materials discovery for energy sustainability. This review examines how auto‐generated high‐quality materials databases and domain‐specific language models accelerate research in photovoltaics, thermoelectrics, batteries and magnetic materials. Applications involve extraction of
Jacqueline M. Cole
wiley   +1 more source

Multimodal Handwritten Exam Text Recognition Based on Deep Learning

Applied Sciences
To address the complex challenge of recognizing mixed handwritten text in practical scenarios such as examination papers and to overcome the limitations of existing methods that typically focus on a single category, this paper proposes MHTR, a Multimodal
Hua Shi, Zhenhui Zhu, Chenxue Zhang, Xiaozhou Feng, Yonghang Wang   +4 more
doaj   +1 more source

On‐Chip Photonic Neural Network Architectures

Advanced Optical Materials, EarlyView.
This review presents a comprehensive overview of on‐chip photonic neural network architectures, covering key photonic building blocks, representative network types, and emerging applications. Recent advances, implementation challenges, and future directions are examined, highlighting the potential of integrated photonics to enable ultrafast, energy ...
Seokjin Hong, Berkay Neseli, Seungyeon Cho, Seungsoo Lee, Yeonjun Kim, Seungyoon Choi, Hyo‐Hoon Park, Hamza Kurt   +7 more
wiley   +1 more source

Backpropagation Through Soft Body: Investigating Information Processing in Brain–Body Coupling Systems

Advanced Robotics Research, EarlyView.
This study explores how information processing is distributed between brains and bodies through a codesign approach. Using the “backpropagation through soft body” framework, brain–body coupling agents are developed and analyzed across several tasks in which output is generated through the agents’ physical dynamics.
Hiroki Tomioka, Katsuma Inoue, Yasuo Kuniyoshi, Kohei Nakajima   +3 more
wiley   +1 more source

Seam carving, horizontal projection profile and contour tracing for line and word segmentation of language independent handwritten documents

Results in Engineering, 2023
Handwritten documents are, as always, highly challenging for recognition tasks compared to printed documents. Rather than using isolated characters as elementary components for recognition, practical documents use words or character strings.
Mamatarani Das, Mrutyunjaya Panda
doaj   +1 more source

Exact Discrete Stochastic Simulation With Deep‐Learning‐Scale Gradient Optimization

Advanced Science, EarlyView.
A 203,796‐parameter gene regulatory network classifies handwritten digits with 98.4% accuracy using exact stochastic dynamics. The framework decouples forward simulation from backward differentiation, making continuous‐time Markov chain models compatible with deep‐learning optimization.
Jose M. G. Vilar, Leonor Saiz
wiley   +1 more source

Enhancement of handwritten text recognition using AI-based hybrid approach

MethodsX
Handwritten text recognition (HTR) within computer vision and image processing stands as a prominent and challenging research domain, holding significant implications for diverse applications.
Supriya Mahadevkar, Shruti Patil, Ketan Kotecha   +2 more
doaj   +1 more source

Lexicon and attention based handwritten text recognition system

Machine Graphics & Vision, 2022
The handwritten text recognition problem is widely studied by the researchers of computer vision community due to its scope of improvement and applicability to daily lives. It is a sub-domain of pattern recognition.
Lalita Kumari, Sukhdeep Singh, Vaibhav Varish Singh Rathore, Anuj Sharma   +3 more
doaj   +1 more source

On the Generalization of Handwritten Text Recognition Models

2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Recent advances in Handwritten Text Recognition (HTR) have led to significant reductions in transcription errors on standard benchmarks under the i.i.d. assumption, thus focusing on minimizing in-distribution (ID) errors. However, this assumption does not hold in real-world applications, which has motivated HTR research to explore Transfer Learning and
Carlos Garrido-Munoz, Jorge Calvo-Zaragoza   +1 more
openaire   +3 more sources

Strategy of Triple‐Gradient in Binary Pixels for Flexible Pressure Sensing with High Sensitivity and Wide‐Range Linearity

Advanced Science, EarlyView.
A flexible pressure sensor with triple‐gradient design of conductivity, modulus, and dimension in binary micro‐dome pixels is proposed. Based on precisely‐designed CNT/PDMS matrix, the device exhibits a linear sensitivity of 974.1 kPa−1 across range up to 1.8 MPa (R2 > 0.99), offering an effective strategy for potential applications in healthcare ...
Yifan Liu, Xiao Xu, Fengming Hu, Yuanzhe Liang, Zidong Yin, Biao Qi, Ruolin Liu, Ziyi Dai, Jianyi Luo, Bingpu Zhou   +9 more
wiley   +1 more source