Ching Y. Suen

Abbas Zohrevand, Zahra Imani, Javad Sadri et al.

Most of the methods on handwritten recognition in the literature are focused and evaluated on Black and White (BW) image databases. In this paper we try to answer a fundamental question in document recognition. Using Convolutional Neural Networks (CNNs), as eye simulator, we investigate to see whether color modalities of handwritten digits and words affect their recognition accuracy or speed? To the best of our knowledge, so far this question has not been answered due to the lack of handwritten databases that have all three color modalities of handwritings. To answer this question, we selected 13,330 isolated digits and 62,500 words from a novel Persian handwritten database, which have three different color modalities and are unique in term of size and variety. Our selected datasets are divided into training, validation, and testing sets. Afterwards, similar conventional CNN models are trained with the training samples. While the experimental results on the testing set show that CNN on the BW digit and word images has a higher performance compared to the other two color modalities, in general there are no significant differences for network accuracy in different color modalities. Also, comparisons of training times in three color modalities show that recognition of handwritten digits and words in BW images using CNN is much more efficient.

Zahra Ebrahimi Vargoorani, Ching Yee Suen

License plate detection (LPD) is essential for traffic management, vehicle tracking, and law enforcement but faces challenges like variable lighting and diverse font types, impacting accuracy. Traditionally reliant on image processing and machine learning, the field is now shifting towards deep learning for its robust performance in various conditions. Current methods, however, often require tailoring to specific regional datasets. This paper proposes a dual deep learning strategy using a Faster R-CNN for detection and a CNN-RNN model with Connectionist Temporal Classification (CTC) loss and a MobileNet V3 backbone for recognition. This approach aims to improve model performance using datasets from Ontario, Quebec, California, and New York State, achieving a recall rate of 92% on the Centre for Pattern Recognition and Machine Intelligence (CENPARMI) dataset and 90% on the UFPR-ALPR dataset. It includes a detailed error analysis to identify the causes of false positives. Additionally, the research examines the role of font features in license plate (LP) recognition, analyzing fonts like Driver Gothic, Dreadnought, California Clarendon, and Zurich Extra Condensed with the OpenALPR system. It discovers significant performance discrepancies influenced by font characteristics, offering insights for future LPD system enhancements. Keywords: Deep Learning, License Plate, Font Evaluation

Wenhao Gu, Li Gu, Ziqiang Wang et al.

Despite recent significant advancements in Handwritten Document Recognition (HDR), the efficient and accurate recognition of text against complex backgrounds, diverse handwriting styles, and varying document layouts remains a practical challenge. Moreover, this issue is seldom addressed in academic research, particularly in scenarios with minimal annotated data available. In this paper, we introduce the DocTTT framework to address these challenges. The key innovation of our approach is that it uses test-time training to adapt the model to each specific input during testing. We propose a novel Meta-Auxiliary learning approach that combines Meta-learning and self-supervised Masked Autoencoder~(MAE). During testing, we adapt the visual representation parameters using a self-supervised MAE loss. During training, we learn the model parameters using a meta-learning framework, so that the model parameters are learned to adapt to a new input effectively. Experimental results show that our proposed method significantly outperforms existing state-of-the-art approaches on benchmark datasets.

Hiba Fallatah, Ching Suen, Olga Ormandjieva

Reliance on anonymity in social media has increased its popularity on these platforms among all ages. The availability of public Wi-Fi networks has facilitated a vast variety of online content, including social media applications. Although anonymity and ease of access can be a convenient means of communication for their users, it is difficult to manage and protect its vulnerable users against sexual predators. Using an automated identification system that can attribute predators to their text would make the solution more attainable. In this survey, we provide a review of the methods of pedophile attribution used in social media platforms. We examine the effect of the size of the suspect set and the length of the text on the task of attribution. Moreover, we review the most-used datasets, features, classification techniques and performance measures for attributing sexual predators. We found that few studies have proposed tools to mitigate the risk of online sexual predators, but none of them can provide suspect attribution. Finally, we list several open research problems.

Xu-Yao Zhang, Cheng-Lin Liu, Ching Y. Suen

The accuracies for many pattern recognition tasks have increased rapidly year by year, achieving or even outperforming human performance. From the perspective of accuracy, pattern recognition seems to be a nearly-solved problem. However, once launched in real applications, the high-accuracy pattern recognition systems may become unstable and unreliable, due to the lack of robustness in open and changing environments. In this paper, we present a comprehensive review of research towards robust pattern recognition from the perspective of breaking three basic and implicit assumptions: closed-world assumption, independent and identically distributed assumption, and clean and big data assumption, which form the foundation of most pattern recognition models. Actually, our brain is robust at learning concepts continually and incrementally, in complex, open and changing environments, with different contexts, modalities and tasks, by showing only a few examples, under weak or noisy supervision. These are the major differences between human intelligence and machine intelligence, which are closely related to the above three assumptions. After witnessing the significant progress in accuracy improvement nowadays, this review paper will enable us to analyze the shortcomings and limitations of current methods and identify future research directions for robust pattern recognition.

Azadeh Nazemi, Niloofar Tavakolian, Donal Fitzpatrick et al.

This paper describes an approach for offline recognition of handwritten mathematical symbols. The process of symbol recognition in this paper includes symbol segmentation and accurate classification for over 300 classes. Many multidimensional mathematical symbols need both horizontal and vertical projection to be segmented. However, some symbols do not permit to be projected and stop segmentation, such as the root symbol. Besides, many mathematical symbols are structurally similar, specifically in handwritten such as 0 and null. There are more than 300 Mathematical symbols. Therefore, designing an accurate classifier for more than 300 classes is required. This paper initially addresses the issue regarding segmentation using Simple Linear Iterative Clustering (SLIC). Experimental results indicate that the accuracy of the designed kNN classifier is 84% for salient, 57% Histogram of Oriented Gradient (HOG), 53% for Linear Binary Pattern (LBP) and finally 43% for pixel intensity of raw image for 66 classes. 87 classes using modified LeNet represents 90% accuracy. Finally, for 101 classes, SqueezeNet ac

Dan Liu, Ching Yee Suen, Olga Ormandjieva

Recurrent Neural Networks with Long Short-Term Memory cell (LSTM-RNN) have impressive ability in sequence data processing, particularly for language model building and text classification. This research proposes the combination of sentiment analysis, new approach of sentence vectors and LSTM-RNN as a novel way for Sexual Predator Identification (SPI). LSTM-RNN language model is applied to generate sentence vectors which are the last hidden states in the language model. Sentence vectors are fed into another LSTM-RNN classifier, so as to capture suspicious conversations. Hidden state enables to generate vectors for sentences never seen before. Fasttext is used to filter the contents of conversations and generate a sentiment score so as to identify potential predators. The experiment achieves a record-breaking accuracy and precision of 100% with recall of 81.10%, exceeding the top-ranked result in the SPI competition.