Multi-Script Text Detection from Image Using FRCNN

The online shopping is increasing rapidly because of its convenience to buy from home and comparing products from their reviews written by other purchasers. When people buy a product, they express their emotions about that product in the form of review. In Indian context, it is found that the reviews contain Hindi text along with English. It is also found that most of the Hindi text contains opinionated words like bahut achha, bakbas, pesa wasool etc. We have tried to find out different Hindi texts appearing in product reviews written on Indian E-commerce portals. We have also developed a system which takes all those reviews containing Hindi as well as English texts and find out the sentiment expressed in that review for each attribute of the product as well as a final review of the product.

Scene texts refer to arbitrary text found in images captured by cameras in real-world settings. The tasks of text detection and recognition are critical components of computer vision, with applications spanning scene understanding, information retrieval, robotics, and autonomous driving. Despite significant advancements in deep learning methods, achieving accurate text detection and recognition in complex images remains a formidable challenge for robust real-world applications. Several factors contribute to these challenges. First, the diversity of text shapes, fonts, colors, and styles complicates detection efforts. Second, the myriad combinations of characters, often with unstable attributes, make complete detection difficult, especially when background interruptions obscure character strokes and shapes. Finally, effective coordination of multiple sub-tasks in end-to-end learning is essential for success. This research aimed to tackle these challenges by enhancing text discriminative representation. This study focused on two interconnected problems: Scene Text Recognition (STR), which involves recognizing text from scene images, and Scene Text Detection (STD), which entails simultaneously detecting and recognizing multiple texts within those images. This research focuses on implementing and evaluating the Efficient and Accurate Scene Text Detector (EAST) algorithm for text detection and recognition in natural scene images. The study aims to compare the performance of three prominent Optical Character Recognition (OCR) techniques—TesseractOCR, PaddleOCR, and EasyOCR. The EAST model was applied to a series of sample test images, and the results were visually represented with bounding boxes highlighting the detected text regions. The inference times for each image were recorded, highlighting the algorithm's efficiency, with average times of 0.446, 0.439, and 0.440 seconds for the respective test images. These results indicate that the EAST algorithm is accurate and operates in real-time, making it suitable for applications requiring immediate text recognition.

In this paper, we present a DAV-based system for text (mainly English and Chinese) detection and recognition. With the combination of unmanned aerial vehicle and scene text recognition, the system can realize text detection and recognition in long-range air-plane images, providing an underlay for unmanned navigation and fast text information understanding. Robust text detection and accurate text recognition can be achieved by two contributions. First, a scalable engine is proposed to synthesize text images by overlaying English or Chinese text into existing images in a natural way. Second, an framework which is trainable and end-to-end by combining Convolutional Neural Network and Recurrent Neural Network is adapted to recognize the variable-length text with a high accuracy. Field experiments are performed with different videos shot in various backgrounds and outdoors to show that the proposed system can detect and recognise text information in UAV imagery robustly and effectively.

In this digital era, people rely on the internet for their news consumption. As people are free to express their opinions on social media, much information shared on the internet is loaded with propaganda. Propagandist contents are intended to influence public opinion. In the mainstream media or prominent news agencies, the authors’ and news agencies’ own bias may impact in the news contents. Hence, it is required to detect such propaganda spread through news articles. Detection and classification of propagandist text require standard, high-quality, annotated datasets. A few datasets are available for propaganda classification. However, these datasets are mostly in English. Hindi is the most spoken language in India, and efforts are needed to detect its propagandist contents. This research work introduces two new datasets: H-Prop and H-Prop-News, which consist of news articles in Hindi annotated as propaganda or non-propaganda. The H-Prop dataset is generated by translating 28,630 news articles from the QProp dataset. The H-Prop-News dataset contains 5500 news articles collected from 32 prominent Hindi news websites. We experiment with the proposed datasets using four supervised machine learning models combined with different feature vectors and word embeddings. Our experiments achieve 87% accuracy using Logistic Regression with TF-IDF feature vectors. The datasets provide high-quality labeled news articles in Hindi and open new avenues for researchers to explore techniques for analyzing and classifying propaganda in Hindi text.

Optical Character Recognition (OCR) is the process of recognizing characters automatically from scanned or image documents. OCR software uses machine learning to recognize characters in the document. Such software needs to pass a training phase to learn how to recognize the letters in the text. In order to implement the training phase the OCR needs to use a standard dataset. The dataset can be used to evaluate the obtained results. In this research, we propose an Arabic printed OCR dataset. To the best of our knowledge, there is no Arabic OCR dataset that is available to be used by the research community with its ground truth with a size that is suitable to build a robust Arabic OCR. The proposed dataset is extracted randomly from Wikipedia to have different topics. It consists of 4,587 Arabic articles with a total of 8,994 images.

Most of the text detection and script classification approaches from natural scenes only cater for a single script whereas text in natural scenes may come in various scripts. This research proposes a gestalt-based approach for multi-script text detection and classification based on human perception. Human perceptual organization is where humans are able to organize visual input into meaningful information. This approach is based on the figure-ground articulation where we perceive the figure or text as standing in front of the background. Features extracted from wavelet coefficients and MSER is used as input to SVM for text detection and script classification. Experimental results indicate that this approach is competitive with the state of the art text detection and script classification approaches.

The social media platform is a convenient medium to express personal thoughts\nand share useful information. It is fast, concise, and has the ability to reach\nmillions. It is an effective place to archive thoughts, share artistic content,\nreceive feedback, promote products, etc. Despite having numerous advantages\nthese platforms have given a boost to hostile posts. Hate speech and derogatory\nremarks are being posted for personal satisfaction or political gain. The\nhostile posts can have a bullying effect rendering the entire platform\nexperience hostile. Therefore detection of hostile posts is important to\nmaintain social media hygiene. The problem is more pronounced languages like\nHindi which are low in resources. In this work, we present approaches for\nhostile text detection in the Hindi language. The proposed approaches are\nevaluated on the Constraint@AAAI 2021 Hindi hostility detection dataset. The\ndataset consists of hostile and non-hostile texts collected from social media\nplatforms. The hostile posts are further segregated into overlapping classes of\nfake, offensive, hate, and defamation. We evaluate a host of deep learning\napproaches based on CNN, LSTM, and BERT for this multi-label classification\nproblem. The pre-trained Hindi fast text word embeddings by IndicNLP and\nFacebook are used in conjunction with CNN and LSTM models. Two variations of\npre-trained multilingual transformer language models mBERT and IndicBERT are\nused. We show that the performance of BERT based models is best. Moreover, CNN\nand LSTM models also perform competitively with BERT based models.\n

The object detection approaches in conjunction with Fast/Faster R-CNN and YOLO have shown the benchmarking performance on several occasions. Inspired by the Refine Net, we propose a new model called Faster+ R-CNN based on Faster R-CNN, which is mainly based on iterative refinement on the proposed regions. The Faster+ R-CNN model can iteratively refine the region proposal based on previous output. We trained and tested our new model on PASCAL VOC 2007 dataset, and experiments showed that our method can iteratively improve the mean average precision (mAP) from 0.6702 to 0.6764 in object detecting task. We also demonstrate the facial detection results using the Faster+ R-CNN on the widely used Face Detection Dataset and Benchmark (FDDB) benchmark. By training the Faster+ R-CNN model on the large scale WIDER face dataset, we report the improved results on two widely used face detection benchmarks including FDDB.

In order to improve communication accessibility for people with speech and hearing impairments, sign language recognition is essential. A 3D Convolutional Neural Network (3D-CNN) is used in this study for sign language recognition system to categorize hand gestures from video clips. The suggested model ensures high accuracy and efficiency by classifying and extracting features from video frames. Several sign gestures are included in the dataset, which has undergone extensive preprocessing methods like frame extraction, augmentation, and normalization to increase robustness. By incorporating multilingual translation capabilities, the system expands its accessibility by translating recognized gestures into text in Hindi, Kannada, and English. The model's efficiency in practical applications, such as assistive communication devices and human-computer interaction, is demonstrated by our experimental results, which show an astounding 99% accuracy. This study aids in the advancement of inclusive AI-powered solutions that help people with hearing loss communicate.

India is a multilingual country with 22 official languages and more than 1600 languages in existence. Kannada is one of the official languages and widely used in the state of Karnataka whose population is over 65 million. Kannada is one of the south Indian languages and it stands in the 33rd position among the list of widely spoken languages across the world. However, the survey reveals that much more effort is required to develop a complete Optical Character Recognition (OCR) system. In this direction the present research work throws light on the development of suitable methodology to achieve the goal of developing an OCR. It is noted that the overall accuracy of the OCR system largely depends on the accuracy of the segmentation phase. So it is desirable to have a robust and efficient segmentation method. In this paper, a method has been proposed for proper segmentation of the text to improve the performance of OCR at the later stages. In the proposed method, the segmentation has been done using horizontal projection profile and windowing. The result obtained is passed to the recognition module. The Histogram of Oriented Gradient (HoG) is used for the recognition in combination with the support vector machine (SVM). The result is taken as the feedback and fed to the segmentation module to improve the accuracy. The experimentation is delivered promising results.

A multilingual document may contain text words in more than one language. In a multilingual country like India it is necessary that a document should be composed of text contents in different languages in order to reach a larger cross section of people, But on the other hand, this causes practical difficulty in OCRing such a document, because the language type of the text should be pre-determined, before employing a particular OCR (Optical Character Recognition). It is perhaps impossible to design a single recognizer which can identify a large number of scripts/languages. So, it is necessary to identify the language region of the document before feeding the document to the corresponding OCR system. Script identification aims to extract information presented in digital documents namely articles, newspapers, magazines and e-books. This has given rise to many language identification systems. The objective of this paper is to propose a model to identify script type of different text portions using visual clues. In this work seven feature namely bottom max row, top horizontal lines, vertical lines, bottom components, tick components, top holes and bottom holes have been used to identify the script type. In this work, multilingual documents with Telugu, English and Hindi scripts have been used. From the experimentation it is understood that the identification accuracy of above 93% is achieved.

This paper describes a new expert system for automatically correcting errors made by optical character recognition (OCR) devices. The system, which we call the post-processing system, is designed to improve the quality of text produced by an OCR device in preparation for subsequent retrieval from an information system. The system is composed of numerous parts: an information retrieval system, an English dictionary, a domain-specific dictionary, and a collection of algorithms and heuristics designed to correct as many OCR errors as possible. For the remaining errors that cannot be corrected, the system passes them on to a user-level editing program. This post-processing system can be viewed as part of a larger system that would streamline the steps of taking a document from its hard copy form to its usable electronic form, or it can be considered a stand alone system for OCR error correction. An earlier version of this system has been used to process approximately 10,000 pages of OCR generated text. Among the OCR errors discovered by this version, about 87% were corrected. We implement numerous new parts of the system, test this new version, and present the results.© (1994) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Objective: We propose a deep-learning-based underwater target detection system that can effectively solve the problem of underwater optical image target detection and recognition. Methods: In this paper, based on the depth of the underwater optical image target detection and recognition and using a learning model, we put forward corresponding solutions using the concept of style migration solutions, such as training samples. A lack of variability and poor generalization of practical applications presents a challenge for underwater object identification. The UW_YOLOv3 lightweight model was proposed to solve the problems of calculating energy consumption and storage resource limitations in underwater application scenarios. The detection and recognition module, based on deep learning, can deal with the degradation process of underwater imaging by embedding an image enhancement module into the detection and recognition module for the joint tuning and transferring of knowledge. Results: The detection accuracy of the UW_YOLOv3 model designed in this paper outperformed the lightweight algorithm YOLOV3-TINY by 7.9% at the same image scale input. Compared with other large algorithms, the detection accuracy was lower, but the detection speed was much higher. Compared with the SSD algorithm, the detection accuracy was only 4.7 lower; the speed was 40.9 FPS higher; and the rate was nearly 16 times higher than Faster R-CNN. When the input scale was 224, although part of the accuracy was lost, the detection speed doubled, reaching 156.9 FPS. Conclusion: Based on our framework, the problem of underwater optical image target detection and recognition can be effectively solved. Relevant studies have not only enriched the theory of target detection and glory, but have also provided optical glasses with a clear vision for appropriate underwater application systems.

Object detection typically assumes that training and test data are drawn from an identical distribution, which, however, does not always hold in practice. Such a distribution mismatch will lead to a significant performance drop. In this work, we aim to improve the cross-domain robustness of object detection. We tackle the domain shift on two levels: (1) the image-level shift, such as image style, illumination, etc., and (2) the instance-level shift, such as object appearance, size, etc. We build our approach based on the recent state-of-the-art Faster R-CNN model, and design two domain adaptation components, on image level and instance level, to reduce the domain discrepancy. The two domain adaptation components are based on <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">H</i> -divergence theory, and are implemented by learning a domain classifier in adversarial training manner. The domain classifiers on different levels are further reinforced with a consistency regularization to learn a domain-invariant region proposal network (RPN) in the Faster R-CNN model. We evaluate our newly proposed approach using multiple datasets including Cityscapes, KITTI, SIM10K, etc. The results demonstrate the effectiveness of our proposed approach for robust object detection in various domain shift scenarios.

With the rapid growth of global maritime traffic, effective ship detection from satellite and aerial images has become essential for applications in maritime security, disaster response, and environmental monitoring. Traditional ship detection methods struggle with the challenges posed by the vastness, complexity, and dynamic nature of oceanic environments. To address these challenges, we propose a highly efficient model that combines the power of CNN with the RPN for fast and accurate ship detection. This study introduces a novel approach for real-time ship object detection in RSI images using the Faster R-CNN. The proposed model is optimized for real-time processing, ensuring the rapid detection of ships even in large-scale satellite imagery. By utilizing a comprehensive dataset of high-resolution RSI images, the system is trained to recognize ships under varying conditions, including different angles, lighting, and sea states. The Faster R-CNN model demonstrates superior performance, offering both high detection accuracy and fast inference times, making it suitable for deployment in real-world maritime surveillance systems. The proposed approach outperformed more conventional R-CNN models in terms of accuracy (99.25 percent for ResNet-50) in experiments conducted on the ship object benchmark dataset.