Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
42,805
result(s) for
"Handwriting"
Sort by:
Abdul's story
Abdul loves telling stories but thinks his messy handwriting and spelling mistakes will keep him from becoming an author, until Mr. Muhammad visits and encourages him to persist.
Book
Reservoir computing using dynamic memristors for temporal information processing
Reservoir computing systems utilize dynamic reservoirs having short-term memory to project features from the temporal inputs into a high-dimensional feature space. A readout function layer can then effectively analyze the projected features for tasks, such as classification and time-series analysis. The system can efficiently compute complex and temporal data with low-training cost, since only the readout function needs to be trained. Here we experimentally implement a reservoir computing system using a dynamic memristor array. We show that the internal ionic dynamic processes of memristors allow the memristor-based reservoir to directly process information in the temporal domain, and demonstrate that even a small hardware system with only 88 memristors can already be used for tasks, such as handwritten digit recognition. The system is also used to experimentally solve a second-order nonlinear task, and can successfully predict the expected output without knowing the form of the original dynamic transfer function.
Reservoir computing facilitates the projection of temporal input signals onto a high-dimensional feature space via a dynamic system, known as the reservoir. Du et al. realise this concept using metal-oxide-based memristors with short-term memory to perform digit recognition tasks and solve non-linear problems.
Journal Article
A self-attention-based deep architecture for online handwriting recognition
The self-attention mechanism has been the most frequent and efficient way for processing and learning sequences in numerous domains of artificial intelligence, including natural language processing, automatic speech recognition, and computer vision in recent years. It has a strong ability to learn the dependencies between the points of the input sequence, particularly those that are separated by a distance, and it also allows for parallel processing of the sequence. As a result, when used in processing sequences, this mechanism has a stronger ability to extract an appropriate representation from the input sequence at a faster rate than other approaches such as recurrent neural networks. Despite the benefits of the self-attention mechanism, recurrent neural networks along with feature engineering have been the most commonly employed approaches to online handwriting recognition. This study introduces an end-to-end online handwriting recognition system that utilizes the self-attention mechanism into three different modeling methods: CTC-based, RNN-T, and encoder–decoder. The proposed system demonstrates the capacity to recognize handwritten scripts without the need for feature engineering. The system’s performance was evaluated using the Arabic Online-KHATT dataset and the English IAM-OnDB dataset. On the former, it achieved character error rate (CER) of 4.78% and word error rate (WER) of 20.63%, and on the latter, the CER of 4.10% and the WER of 14.31%, both of which were noticeably better than the results previously reported. Additionally, the Persian Online Handwriting Database was utilized for experimental validation, resulting in a CER 8.03% and a WER of 28.39%.
Journal Article
SiGe epitaxial memory for neuromorphic computing with reproducible high performance based on engineered dislocations
Although several types of architecture combining memory cells and transistors have been used to demonstrate artificial synaptic arrays, they usually present limited scalability and high power consumption. Transistor-free analog switching devices may overcome these limitations, yet the typical switching process they rely on—formation of filaments in an amorphous medium—is not easily controlled and hence hampers the spatial and temporal reproducibility of the performance. Here, we demonstrate analog resistive switching devices that possess desired characteristics for neuromorphic computing networks with minimal performance variations using a single-crystalline SiGe layer epitaxially grown on Si as a switching medium. Such epitaxial random access memories utilize threading dislocations in SiGe to confine metal filaments in a defined, one-dimensional channel. This confinement results in drastically enhanced switching uniformity and long retention/high endurance with a high analog on/off ratio. Simulations using the MNIST handwritten recognition data set prove that epitaxial random access memories can operate with an online learning accuracy of 95.1%.
Journal Article
HWOBC-A Handwriting Oracle Bone Character Recognition Database
The oracle bone character (OBC) from ancient China is the most famous ancient writing systems around the world. Identifying and deciphering OBCs is one of the most important topics in oracle bone study. In research, one of the challenges is that the literature review usually leads to a huge cost of time and manpower. Therefore, the digitazation of OBC literature through the automatic recognition is the inevitable trend of future development. However, the OBCs in the literature are usually writing characters while the database of handwriting OBC has not yet been presented. In this paper, we establish a handwriting oracle bone character database called HWOBC, containing 83,245 character-level samples which are grouped into 3881-character categories. We also present the performance of several baseline DCNN-based methods, in which Melnyk-Net exhibits the best accuracy of 97.64%. It is anticipated that the publication of this database will facilitate the development of OBC research.
Journal Article
Validity of kinematics measures to assess handwriting development and disorders with a graphomotor task
IntroductionHandwriting disorder is considered to be one of the major public health problems among school-aged children worldwide. All the scales in the literature use handwriting tasks but it could be interesting to investigate a more accurate assessment of handwriting difficulties before the development and acquisition of handwriting as such.ObjectivesThe objective of our study is to examine the validity of a prescriptural task consisting of copying a line of cycloid loops in the diagnosis of handwriting disorders.Methods35 children with handwriting disabilities and 331 typically developing right-handed children in primary school, aged 6-11 years old, were included in the study. They performed a copy of a line of cycloid loops, in an ecological setting, with a paper sheet put on the table. The kinematic measures were recorded with a digital pen. A Receiver Operating Characteristic method (ROC curve) was used to determine whether the loops line copy may be a sensitive test to diagnose handwriting disorders.ResultsSix kinematic variables recorded during the prescriptural task were found to be relevant markers of handwriting disorders with a sensibility between 0.743 and 0.880: strokes number, total and effective drawing time, in-air pauses times, loops number, number of peaks velocity.ConclusionsThe graphomotor task of copying a line of cycloid loops showed a good sensitivity to diagnose handwriting disorders and appeared to be a good predictor test, more particularly with the variables reflecting the strokes temporal organization.Drawing loops is a rapid graphomotor task, useful for exploring prerequisites of handwriting in screening for handwriting disorders.
Journal Article
Importance and challenges of handwriting recognition with the implementation of machine learning techniques: a survey
Ancient manuscripts store historical, literary, cultural, and geographical information. Therefore, the automatic analysis of manuscripts is of great interest in heritage culture and history preservation. Different approaches to handwriting recognition using images have been applied to analyze manuscripts. However, reliable handwriting recognition is a considerable challenge due to different factors related to the writer, the design, the script, the manuscript, and the economy. This paper presents the most relevant works in handwriting recognition using machine learning techniques. The contributions are: i) provide a review of previous research addressing handwriting recognition, ii) depict the general methodology using machine learning in handwriting recognition, iii) highlight relevant works at different levels of analysis (character, word, text line, and text block), iv) present handwriting datasets including the type of content they have, script and language, and v) present the importance and challenges in handwriting recognition. We are confident that the insights and reflections from this review will have a positive impact on the gaps for future research in handwriting recognition.
Journal Article
Advanced handwriting identification: Triboelectric sensor array integrating with deep learning toward high information security
Handwriting identification is widely accepted as scientific evidence. However, its authenticity is questioned because it depends on the appraiser's professional skills and susceptibility to deliberate false identification by expert witnesses. Consequently, there is an urgent need for an effective handwriting identification system (HWIS) that reduces reliance on the appraiser's skills and mitigates the risk of international false identification. Here, we report a HWIS that integrates a self‐powered handwriting signal data acquisition device with an advanced deep learning architecture possessing powerful feature extraction ability and one‐class classification function. The device successfully captures the characteristic differences in handwriting behavior between genuine writers and forgers, and the handwriting identification results demonstrate the excellent performance of our system, showcasing its powerful potential to solve the longstanding challenge of handwriting identification that has perplexed humans for a considerable period. Moreover, this work exhibits the system's capability for remote access and downloading the handwriting signal data through the data cloud, highlighting its practical value for fulfilling the requirements of handwriting recognition and identification applications, and it can effectively advance signature information security and ensure the protection of private information. A handwriting identification system is established by integrating a self‐powered handwriting signal acquisition device with an advanced deep‐learning architecture. The system demonstrated excellent performance in identifying the handwriting of genuine writers and forgers, which has perplexed humans for a considerable period. This work also showcases the system's ability to handle various language character types, effectively advancing signature information security.
Journal Article
Convolutional Neural Network Based Intelligent Handwritten Document Recognition
This paper presents a handwritten document recognition system based on the convolutional neural network technique. In today’s world, handwritten document recognition is rapidly attaining the attention of researchers due to its promising behavior as assisting technology for visually impaired users. This technology is also helpful for the automatic data entry system. In the proposed system prepared a dataset of English language handwritten character images. The proposed system has been trained for the large set of sample data and tested on the sample images of user-defined handwritten documents. In this research, multiple experiments get very worthy recognition results. The proposed system will first perform image pre-processing stages to prepare data for training using a convolutional neural network. After this processing, the input document is segmented using line, word and character segmentation. The proposed system get the accuracy during the character segmentation up to 86%. Then these segmented characters are sent to a convolutional neural network for their recognition. The recognition and segmentation technique proposed in this paper is providing the most acceptable accurate results on a given dataset. The proposed work approaches to the accuracy of the result during convolutional neural network training up to 93%, and for validation that accuracy slightly decreases with 90.42%.
Journal Article
IAMonSense: multi-level handwriting classification using spatiotemporal information
Online handwriting classification has become an open research problem as it serves as a preliminary step for handwriting recognition systems and applications in several other fields. This paper aims to extend the current trends and knowledge with multiple contributions in handwriting classification using spatiotemporal information. Firstly, it enriches the annotations of several publicly available online handwriting datasets, SenseThePen, IAM-onDB, and IAMonDo, for online handwriting classification and recognition tasks. The said datasets are updated with three distinguished levels of annotations, i.e., stroke, sequence, and line levels. The enriched annotations of these datasets extend their functionality for online handwriting classification at different levels for further research analysis. In addition to enrichment, it also unifies the annotation levels across the datasets, which enables the research community to benchmark proposed methods for comparative analysis using multiple datasets. All the datasets with enriched annotations are made publicly available for the research community as part of the IAMonSense dataset. Moreover, this paper presents a comprehensive benchmark of these datasets using multiple deep neural networks such as traditional convolutional neural networks (CNNs), graph convolutional networks(GCNs), attention-based neural networks, and transformers. These benchmarks can be used later on for further development in online handwriting classification.
Journal Article