Search results for: scene text recognition

Authors: Americo Pereira, Ricardo Carvalho, Pedro Carvalho, Luis Corte-Real

Abstract:

Generating 3D content is a task mostly done by hand. It requires specific knowledge not only on how to use the tools for the task but also on the fundamentals of a 3D environment. In this work, we show that automatic generation of content can be achieved, from a scene script, by leveraging existing tools so that non-experts can easily engage in a 3D content generation without requiring vast amounts of time in exploring and learning how to use specific tools. This proposal carries several benefits, including flexible scene synthesis with different levels of detail. Our preliminary results show that the automatically generated content is comparable to the content generated by users with low experience in 3D modeling while vastly reducing the amount of time required for the generation and adds support to implement flexible scenarios for visual scene visualization.

Keywords: 3D virtualization, multimedia, scene script, synthesis

Procedia PDF Downloads 230

Authors: Shuo Liu

Abstract:

In Chinese text classification tasks, redundant words and phrases can interfere with the formation of extracted and analyzed text information, leading to a decrease in the accuracy of the classification model. To reduce irrelevant elements, extract and utilize text content information more efficiently and improve the accuracy of text classification models. In this paper, the text in the corpus is first extracted using the TextRank algorithm for abstraction, the words in the abstract are used as nodes to construct a text graph, and then the graph attention network (GAT) is used to complete the task of classifying the text. Testing on a Chinese dataset from the network, the classification accuracy was improved over the direct method of generating graph structures using text.

Keywords: Chinese natural language processing, text classification, abstract extraction, graph attention network

Procedia PDF Downloads 62

Authors: Samabia Tehsin, Sumaira Kausar

Abstract:

Due to the vast increase in the multimedia data in recent years, efficient and robust retrieval techniques are needed to retrieve and index images/ videos. Text embedded in the images can serve as the strong retrieval tool for images. This is the reason that text extraction is an area of research with increasing attention. English text extraction is the focus of many researchers but very less work has been done on other languages like Urdu. This paper is focusing on Urdu text extraction from video frames. This paper presents a text detection feature set, which has the ability to deal up with most of the problems connected with the text extraction process. To test the validity of the method, it is tested on Urdu news dataset, which gives promising results.

Keywords: caption text, content-based image retrieval, document analysis, text extraction

Procedia PDF Downloads 476

Authors: Shuo Liu, Dan Liu

Abstract:

In the field of named entity recognition, most of the research has been conducted around simple entities. However, for nested named entities, which still contain entities within entities, it has been difficult to identify them accurately due to their boundary ambiguity. In this paper, a hierarchical recognition model is constructed based on the grammatical structure and semantic features of Chinese text for boundary calculation based on lexical features. The analysis is carried out at different levels in terms of granularity, semantics, and lexicality, respectively, avoiding repetitive work to reduce computational effort and using the semantic features of words to calculate the boundaries of entities to improve the accuracy of the recognition work. The results of the experiments carried out on web-based microblogging data show that the model achieves an accuracy of 86.33% and an F1 value of 89.27% in recognizing nested named entities, making up for the shortcomings of some previous recognition models and improving the efficiency of recognition of nested named entities.

Keywords: coarse-grained, nested named entity, Chinese natural language processing, word embedding, T-SNE dimensionality reduction algorithm

Procedia PDF Downloads 96

Authors: Jie Zhao, Meng Su

Abstract:

Image recognition, as one of the most critical technologies in computer vision, works to help machine-like robotics understand a scene, that is, if deployed appropriately, will trigger the revolution in remote sensing and industry automation. With the developments of AI technologies, there are many prevailing and sophisticated neural networks as technologies developed for image recognition. However, computer vision platforms as hardware, supporting neural networks for image recognition, as crucial as the neural network technologies, need to be more congruently addressed as the research subjects. In contrast, different computer vision platforms are deterministic to leverage the performance of different neural networks for recognition. In this paper, three different computer vision platforms – Jetson Nano(with 4GB), a standalone laptop(with RTX 3000s, using CUDA), and Google Colab (web-based, using GPU) are explored and four prominent neural network architectures (including AlexNet, VGG(16/19), GoogleNet, and ResNet(18/34/50)), are investigated. In the context of pairwise usage between different computer vision platforms and distinctive neural networks, with the merits of recognition accuracy and time efficiency, the performances are evaluated. In the case study using public imageNets, our findings provide a nuanced perspective on optimizing image recognition tasks across Edge-AI platforms, offering guidance on selecting appropriate neural network structures to maximize performance under hardware constraints.

Keywords: alexNet, VGG, googleNet, resNet, Jetson nano, CUDA, COCO-NET, cifar10, imageNet large scale visual recognition challenge (ILSVRC), google colab

Procedia PDF Downloads 48

Authors: Vishwanath Pethri Kamath, Jayantha Gowda Sarapanahalli, Vishal Mishra, Siddhesh Balwant Bandgar

Abstract:

Recognition of emotional information is essential in any form of communication. Growing HCI (Human-Computer Interaction) in recent times indicates the importance of understanding of emotions expressed and becomes crucial for improving the system or the interaction itself. In this research work, textual data for emotion recognition is used. The text being the least expressive amongst the multimodal resources poses various challenges such as contextual information and also sequential nature of the language construction. In this research work, the proposal is made for a neural architecture to resolve not less than 8 emotions from textual data sources derived from multiple datasets using google pre-trained word2vec word embeddings and a Multi-head attention-based bidirectional LSTM model with a one-vs-all Multi-Level Classification. The emotions targeted in this research are Anger, Disgust, Fear, Guilt, Joy, Sadness, Shame, and Surprise. Textual data from multiple datasets were used for this research work such as ISEAR, Go Emotions, Affect datasets for creating the emotions’ dataset. Data samples overlap or conflicts were considered with careful preprocessing. Our results show a significant improvement with the modeling architecture and as good as 10 points improvement in recognizing some emotions.

Keywords: text emotion recognition, bidirectional LSTM, multi-head attention, multi-level classification, google word2vec word embeddings

Procedia PDF Downloads 145

Authors: Sandhya Baskaran, Hari Kumar Nagabushanam

Abstract:

Sarcasm sentiment recognition is an area of natural language processing that is being probed into in the recent times. Even with the advancements in NLP, typical translations of words, sentences in its context fail to provide the exact information on a sentiment or emotion of a user. For example, if something bad happens, the statement ‘That's just what I need, great! Terrific!’ is expressed in a sarcastic tone which could be misread as a positive sign by any text-based analyzer. In this paper, we are presenting a unique real time ‘word with its tone’ spotting technique which would provide the sentiment analysis for a tone or pitch of a voice in combination with the words being expressed. This hybrid approach increases the probability for identification of special sentiment like sarcasm much closer to the real world than by mining text or speech individually. The system uses a tone analyzer such as YIN-FFT which extracts pitch segment-wise that would be used in parallel with a speech recognition system. The clustered data is classified for sentiments and sarcasm score for each of it determined. Our Simulations demonstrates the improvement in f-measure of around 12% compared to existing detection techniques with increased precision and recall.

Keywords: sarcasm recognition, tone-word spotting, natural language processing, pitch analyzer

Procedia PDF Downloads 258

Authors: Yanjun Chen, Chuanping Hu, Jie Shao, Lin Mei, Chongyang Zhang

Abstract:

Traditional image classification is based on encoding scheme (e.g. Fisher Vector, Vector of Locally Aggregated Descriptor) with low-level image features (e.g. SIFT, HoG). Compared to these low-level local features, deep convolutional features obtained at the mid-level layer of convolutional neural networks (CNN) have richer information but lack of geometric invariance. For scene classification, there are scattered objects with different size, category, layout, number and so on. It is crucial to find the distinctive objects in scene as well as their co-occurrence relationship. In this paper, we propose a method to take advantage of both deep convolutional features and the traditional encoding scheme while taking object-centric and scene-centric information into consideration. First, to exploit the object-centric and scene-centric information, two CNNs that trained on ImageNet and Places dataset separately are used as the pre-trained models to extract deep convolutional features at multiple scales. This produces dense local activations. By analyzing the performance of different CNNs at multiple scales, it is found that each CNN works better in different scale ranges. A scale-wise CNN adaption is reasonable since objects in scene are at its own specific scale. Second, a fisher kernel is applied to aggregate a global representation at each scale and then to merge into a single vector by using a post-processing method called scale-wise normalization. The essence of Fisher Vector lies on the accumulation of the first and second order differences. Hence, the scale-wise normalization followed by average pooling would balance the influence of each scale since different amount of features are extracted. Third, the Fisher vector representation based on the deep convolutional features is followed by a linear Supported Vector Machine, which is a simple yet efficient way to classify the scene categories. Experimental results show that the scale-specific feature extraction and normalization with CNNs trained on object-centric and scene-centric datasets can boost the results from 74.03% up to 79.43% on MIT Indoor67 when only two scales are used (compared to results at single scale). The result is comparable to state-of-art performance which proves that the representation can be applied to other visual recognition tasks.

Keywords: deep convolutional features, Fisher Vector, multiple scales, scale-specific normalization

Procedia PDF Downloads 299

Authors: Wided Bakari, Patrice Bellot, Omar Trigui, Mahmoud Neji

Abstract:

This article constitutes an opening to think of the modeling and analysis of Arabic texts in the context of a question-answer system. It is a question of exceeding the traditional approaches focused on morphosyntactic approaches. Furthermore, we present a new approach that analyze a text in order to extract correct answers then transform it to logical predicates. In addition, we would like to represent different levels of information within a text to answer a question and choose an answer among several proposed. To do so, we transform both the question and the text into logical forms. Then, we try to recognize all entailment between them. The results of recognizing the entailment are a set of text sentences that can implicate the user’s question. Our work is now concentrated on an implementation step in order to develop a system of question-answering in Arabic using techniques to recognize textual implications. In this context, the extraction of text features (keywords, named entities, and relationships that link them) is actually considered the first step in our process of text modeling. The second one is the use of techniques of textual implication that relies on the notion of inference and logic representation to extract candidate answers. The last step is the extraction and selection of the desired answer.

Keywords: NLP, Arabic language, question-answering, recognition text entailment, logic forms

Procedia PDF Downloads 309

Authors: Kabil Boukhari

Abstract:

In the context of information retrieval, the selection of the most relevant words is a very important step. In fact, the text cleaning allows keeping only the most representative words for a better use. In this paper, we propose a library for the purpose text preprocessing within an implemented application to facilitate this task. This study has two purposes. The first, is to present the related work of the various steps involved in text preprocessing, presenting the segmentation, stemming and lemmatization algorithms that could be efficient in the rest of study. The second, is to implement a developed tool for text preprocessing in French and English. This library accepts unstructured text as input and provides the preprocessed text as output, based on a set of rules and on a base of stop words for both languages. The proposed library has been made on different corpora and gave an interesting result.

Keywords: text preprocessing, segmentation, knowledge extraction, normalization, text generation, information retrieval

Procedia PDF Downloads 58

Authors: E. Calil, L. A. Pereira

Abstract:

The acquisition of orthography is a complex process, involving both lexical and grammatical questions. This learning occurs simultaneously with the domain of multiple textual aspects (e.g.: graphs, punctuation, etc.). However, most of the research on orthographic acquisition focus on this acquisition from an autonomous point of view, separated from the process of textual production. This means that their object of analysis is the production of words selected by the researcher or the requested sentences in an experimental and controlled setting. In addition, the analysis of the Spelling Problems (SP) are identified by the researcher on the sheet of paper. Considering the perspective of Textual Genetics, from an enunciative approach, this study will discuss the SPs recognized by dyads of newly literate students, while they are writing a text collaboratively. Six proposals of textual production were registered, requested by a 2nd year teacher of a Portuguese Primary School between January and March 2015. In our case study we discuss the SPs recognized by the dyad B and L (7 years old). We adopted as a methodological tool the Ramos System audiovisual record. This system allows real-time capture of the text in process and of the face-to-face dialogue between both students and their teacher, and also captures the body movements and facial expressions of the participants during textual production proposals in the classroom. In these ecological conditions of multimodal registration of collaborative writing, we could identify the emergence of SP in two dimensions: i. In the product (finished text): SP identification without recursive graphic marks (without erasures) and the identification of SPs with erasures, indicating the recognition of SP by the student; ii. In the process (text in progress): identification of comments made by students about recognized SPs. Given this, we’ve analyzed the comments on identified SPs during the text in progress. These comments characterize a type of reformulation referred to as Commented Oral Erasure (COE). The COE has two enunciative forms: Simple Comment (SC) such as ' 'X' is written with 'Y' '; or Unfolded Comment (UC), such as ' 'X' is written with 'Y' because...'. The spelling COE may also occur before or during the SP (Early Spelling Recognition - ESR) or after the SP has been entered (Later Spelling Recognition - LSR). There were 631 words entered in the 6 stories written by the B-L dyad, 145 of them containing some type of SP. During the text in progress, the students recognized orally 174 SP, 46 of which were identified in advance (ESRs) and 128 were identified later (LSPs). If we consider that the 88 erasure SPs in the product indicate some form of SP recognition, we can observe that there were twice as many SPs recognized orally. The ESR was characterized by SC when students asked their colleague or teacher how to spell a given word. The LSR presented predominantly UC, verbalizing meta-orthographic arguments, mostly made by L. These results indicate that writing in dyad is an important didactic strategy for the promotion of metalinguistic reflection, favoring the learning of spelling.

Keywords: collaborative writing, erasure, learning, metalinguistic awareness, spelling, text production

Procedia PDF Downloads 138

Authors: Ahmad Alwosheel, Ahmed Alqaraawi

Abstract:

This work proposes an unsupervised text-independent probabilistic approach to recognize Quran reciter voice. It is an accurate approach that works on real time applications. This approach does not require a prior information about reciter models. It has two phases, where in the training phase the reciters' acoustical features are modeled using Gaussian Mixture Models, while in the testing phase, unlabeled reciter's acoustical features are examined among GMM models. Using this approach, a high accuracy results are achieved with efficient computation time process.

Keywords: Quran, speaker recognition, reciter recognition, Gaussian Mixture Model

Procedia PDF Downloads 349

Authors: Ravneet Kaur

Abstract:

Character recognition is a very interesting area of pattern recognition. From past few decades, an intensive research on character recognition for Roman, Chinese, and Japanese and Indian scripts have been reported. In this paper, a review of Handwritten Character Recognition work on Indian Script Gurmukhi is being highlighted. Most of the published papers were summarized, various methodologies were analysed and their results are reported.

Keywords: Gurmukhi character recognition, online, offline, HCR survey

Procedia PDF Downloads 394

Authors: Hamsini Pulugurtha, P. V. S. L. Jagadamba

Abstract:

Programmed Speech to Text and Text Summarization Using Graph-based Algorithms can be utilized in gatherings to get the short depiction of the gathering for future reference. This gives signature check utilizing Siamese neural organization to confirm the personality of the client and convert the client gave sound record which is in English into English text utilizing the discourse acknowledgment bundle given in python. At times just the outline of the gathering is required, the answer for this text rundown. Thus, the record is then summed up utilizing the regular language preparing approaches, for example, solo extractive text outline calculations

Keywords: Siamese neural network, English speech, English text, natural language processing, unsupervised extractive text summarization

Procedia PDF Downloads 179

Authors: Guezouli Larbi, Belkacem Soundes

Abstract:

The automation of the text detection process can help the human in his driving task. Its application can be very useful to help drivers to have more information about their environment by facilitating the reading of road signs such as directional signs, events, stores, etc. In this paper, a system consisting of two stages has been proposed. In the first one, we used pseudo-Zernike moments to pinpoint areas of the image that may contain text. The architecture of this part is based on three main steps, region of interest (ROI) detection, text localization, and non-text region filtering. Then, in the second step, we present a convolutional neural network architecture (On-Road Text Detection Network - ORTDN) which is considered a classification phase. The results show that the proposed framework achieved ≈ 35 fps and an mAP of ≈ 90%, thus a low computational time with competitive accuracy.

Keywords: text detection, CNN, PZM, deep learning

Procedia PDF Downloads 55

Authors: Benish Chaudhry

Abstract:

Most of the accidents these days are occurring because of the ‘text-and-drive’ concept. If we look at the structure of cities in UAE, there are great distances, because of which it is impossible to drive without using or merely checking the cellphone. Moreover, if we look at the road structure, it is almost impossible to stop at a point and text. With the introduction of TEXT STOPs, drivers will be able to stop different stops for a maximum of 1 and a half-minute in order to reply or write a message. They can be introduced at a distance of 10 minutes of driving on the average speed of the road, so the drivers can look forward to a stop and can reply to a text when needed. A user survey indicates that drivers are willing to NOT text-and-drive if they have such a facility available.

Keywords: transport, accidents, urban planning, road planning

Procedia PDF Downloads 345

Authors: Lakshay Kharbanda, Aabhas Chauhan

Abstract:

In this paper, an application of Normalized Compression Distance (NCD) to detect notable scene alterations occurring in videos is presented. Several research groups have been developing methods to perform image classification using NCD, a computable approximation to Normalized Information Distance (NID) by studying the degree of similarity in images. The timeframes where significant aberrations between the frames of a video have occurred have been identified by obtaining a threshold NCD value, using two compressors: LZMA and BZIP2 and defining scene alterations using Pixel Difference Percentage metrics.

Keywords: image compression, Kolmogorov complexity, normalized compression distance, root mean square error

Procedia PDF Downloads 303

Authors: Chien-Hung Chen, Yan-Chun Hsing, Yung-Chun Chang

Abstract:

The Internet has grown into a powerful medium for information dispersion and social interaction that leads to a rapid growth of social media which allows users to easily post their emotions and perspectives regarding certain topics online. Our research aims at using natural language processing and text mining techniques to explore the public emotions expressed on Twitter by analyzing the sentiment behind tweets. In this paper, we propose a composite kernel method that integrates tree kernel with the linear kernel to simultaneously exploit both the tree representation and the distributed emotion keyword representation to analyze the syntactic and content information in tweets. The experiment results demonstrate that our method can effectively detect public emotion of tweets while outperforming the other compared methods.

Keywords: emotion recognition, natural language processing, composite kernel, sentiment analysis, text mining

Procedia PDF Downloads 190

Authors: Vaidehi Murarka, Sneha Mehta, Dishant Upadhyay

Abstract:

Our project topic is “Alphabet Recognition using pixel probability distribution”. The project uses techniques of Image Processing and Machine Learning in Computer Vision. Alphabet recognition is the mechanical or electronic translation of scanned images of handwritten, typewritten or printed text into machine-encoded text. It is widely used to convert books and documents into electronic files etc. Alphabet Recognition based OCR application is sometimes used in signature recognition which is used in bank and other high security buildings. One of the popular mobile applications includes reading a visiting card and directly storing it to the contacts. OCR's are known to be used in radar systems for reading speeders license plates and lots of other things. The implementation of our project has been done using Visual Studio and Open CV (Open Source Computer Vision). Our algorithm is based on Neural Networks (machine learning). The project was implemented in three modules: (1) Training: This module aims “Database Generation”. Database was generated using two methods: (a) Run-time generation included database generation at compilation time using inbuilt fonts of OpenCV library. Human intervention is not necessary for generating this database. (b) Contour–detection: ‘jpeg’ template containing different fonts of an alphabet is converted to the weighted matrix using specialized functions (contour detection and blob detection) of OpenCV. The main advantage of this type of database generation is that the algorithm becomes self-learning and the final database requires little memory to be stored (119kb precisely). (2) Preprocessing: Input image is pre-processed using image processing concepts such as adaptive thresholding, binarizing, dilating etc. and is made ready for segmentation. “Segmentation” includes extraction of lines, words, and letters from the processed text image. (3) Testing and prediction: The extracted letters are classified and predicted using the neural networks algorithm. The algorithm recognizes an alphabet based on certain mathematical parameters calculated using the database and weight matrix of the segmented image.

Keywords: contour-detection, neural networks, pre-processing, recognition coefficient, runtime-template generation, segmentation, weight matrix

Procedia PDF Downloads 355

Authors: Zhu-Qing Jia, Tao Lin, Tong Zhou

Abstract:

The recognition method of Optical Character Recognition has been expensively utilized, while it is rare to be employed specifically in recognition of electronic components. This paper suggests a high-effective algorithm on appearance identification of integrated circuit components based on the existing methods of character recognition, and analyze the pros and cons.

Keywords: optical character recognition, fuzzy page identification, mutual correlation matrix, confidence self-adaptation

Procedia PDF Downloads 507

Authors: Po-Jen Chen, Jian-Jiun Ding, Hung-Wei Hsu, Chien-Yao Wang, Jia-Ching Wang

Abstract:

A more accurate scene classification algorithm using label relations and the hierarchy neural network was developed in this work. In many classification algorithms, it is assumed that the labels are mutually exclusive. This assumption is true in some specific problems, however, for scene classification, the assumption is not reasonable. Because there are a variety of objects with a photo image, it is more practical to assign multiple labels for an image. In this paper, two label relations, which are exclusive relation and hierarchical relation, were adopted in the classification process to achieve more accurate multiple label classification results. Moreover, the hierarchy neural network (hierarchy NN) is applied to classify the image and the directed acyclic graph structure is used for predicting a more reasonable result which obey exclusive and hierarchical relations. Simulations show that, with these techniques, a much more accurate scene classification result can be achieved.

Keywords: convolutional neural network, label relation, hierarchy neural network, scene classification

Procedia PDF Downloads 422

Authors: Victor Ananth Paramananth, Rashid Muniginin, Mahaya Abd Rahman, Siti Afifah Ismail

Abstract:

Safety issues, scene protection, and appropriate evidence collection must be handled in any bio crime scene. There will be a scene or multi-scene to be cordoned for investigation in any bio-incident or bio crime event. Evidence collection is critical in determining the type of microbial or toxin, its lethality, and its source. As a consequence, from the start of the investigation, a proper sampling method is required. The most significant challenges for the crime scene officer would be deciding where to obtain samples, the best sampling method, and the sample sizes needed. Since there could be evidence in liquid, viscous, or powder shape at a crime scene, crime scene officers have difficulty determining which tools to use for sampling. To maximize sample collection, the appropriate tools for sampling methods are necessary. This study aims to assist the crime scene officer in collecting liquid, viscous, and powder biological samples in sufficient quantity while preserving sample quality. Observational tests on sample collection using liquid, viscous, and powder samples for adequate quantity and sample quality were performed using UV light in this research. The density of the light emission varies upon the method of collection and sample types. The best tools for collecting sufficient amounts of liquid, viscous, and powdered samples can be identified by observing UV light. Instead of active microorganisms, the invisible powder is used to assess sufficient sample collection during a crime scene investigation using various collection tools. The liquid, powdered and viscous samples collected using different tools were analyzed using Fourier transform infrared - attenuate total reflection (FTIR-ATR). FTIR spectroscopy is commonly used for rapid discrimination, classification, and identification of intact microbial cells. The liquid, viscous and powdered samples collected using various tools have been successfully observed using UV light. Furthermore, FTIR-ATR analysis showed that collected samples are sufficient in quantity while preserving their quality.

Keywords: biological sample, crime scene, collection tool, UV light, forensic

Procedia PDF Downloads 168

Authors: Faisal Alshuwaier, Ali Areshey

Abstract:

Text mining techniques are generally applied for classifying the text, finding fuzzy relations and structures in data sets. This research provides plenty text mining capabilities. One common application is text classification and event extraction, which encompass deducing specific knowledge concerning incidents referred to in texts. The main contribution of this paper is the clarification of a concept graph generation mechanism, which is based on a text classification and optimal fuzzy relationship extraction. Furthermore, the work presented in this paper explains the application of fuzzy relationship extraction and branch and bound method to simplify the texts.

Keywords: extraction, max-prod, fuzzy relations, text mining, memberships, classification, memberships, classification

Procedia PDF Downloads 545

Authors: Tetyana Baydyk, Ernst Kussul, Sandra Bonilla Meza

Abstract:

There are many articles that attempt to establish the role of different facial fragments in face recognition. Various approaches are used to estimate this role. Frequently, authors calculate the entropy corresponding to the fragment. This approach can only give approximate estimation. In this paper, we propose to use a more direct measure of the importance of different fragments for face recognition. We propose to select a recognition method and a face database and experimentally investigate the recognition rate using different fragments of faces. We present two such experiments in the paper. We selected the PCNC neural classifier as a method for face recognition and parts of the LFW (Labeled Faces in the Wild) face database as training and testing sets. The recognition rate of the best experiment is comparable with the recognition rate obtained using the whole face.

Keywords: face recognition, labeled faces in the wild (LFW) database, random local descriptor (RLD), random features

Procedia PDF Downloads 328

Authors: L. Hamsaveni, Navya Prakash, Suresha

Abstract:

Document Image Analysis recognizes text and graphics in documents acquired as images. An approach without Optical Character Recognition (OCR) for degraded document image analysis has been adopted in this paper. The technique involves document imaging methods such as Image Fusing and Speeded Up Robust Features (SURF) Detection to identify and extract the degraded regions from a set of document images to obtain an original document with complete information. In case, degraded document image captured is skewed, it has to be straightened (deskew) to perform further process. A special format of image storing known as YCbCr is used as a tool to convert the Grayscale image to RGB image format. The presented algorithm is tested on various types of degraded documents such as printed documents, handwritten documents, old script documents and handwritten image sketches in documents. The purpose of this research is to obtain an original document for a given set of degraded documents of the same source.

Keywords: grayscale image format, image fusing, RGB image format, SURF detection, YCbCr image format

Procedia PDF Downloads 341

Authors: Shaima Almalhan, Lubna Shukri, Miriam Talal, Safaa Teskieh

Abstract:

Since artificial intelligence is a rapidly evolving topic that has had a significant impact on technical growth and innovation, this paper examines people's awareness, use, and engagement with the Google Translate application. To see how familiar aware users are with the app and its features, quantitative and qualitative research was conducted. The findings revealed that consumers have a high level of confidence in the application and how far people they benefit from this sort of innovation and how convenient it makes communication.

Keywords: artificial intelligence, google translate, speech recognition, language translation, camera translation, speech to text, text to speech

Procedia PDF Downloads 118

Authors: Yufen Qin

Abstract:

Text sentiment transfer is an important branch of text style transfer. The goal is to generate text with another sentiment attribute based on a text with a specific sentiment attribute while maintaining the content and semantic information unrelated to sentiment unchanged in the process. There are currently two main challenges in this field: no parallel corpus and text attribute entanglement. In response to the above problems, this paper proposed a novel solution: Mask-Prompt-Rerank. Use the method of masking the sentiment words and then using prompt regeneration to transfer the sentence sentiment. Experiments on two sentiment benchmark datasets and one formality transfer benchmark dataset show that this approach makes the performance of small pre-trained language models comparable to that of the most advanced large models, while consuming two orders of magnitude less computing and memory.

Keywords: language model, natural language processing, prompt, text sentiment transfer

Procedia PDF Downloads 45

Authors: S. Tehsin, S. Kausar

Abstract:

The ultimate aim of machine intelligence is to explore and materialize the human capabilities, one of which is the ability to detect various text objects within one or more images displayed on any canvas including prints, videos or electronic displays. Multimedia data has increased rapidly in past years. Textual information present in multimedia contains important information about the image/video content. However, it needs to technologically testify the commonly used human intelligence of detecting and differentiating the text within an image, for computers. Hence in this paper feature set based on anatomical study of human text detection system is proposed. Subsequent examination bears testimony to the fact that the features extracted proved instrumental to text detection.

Keywords: biologically inspired vision, content based retrieval, document analysis, text extraction

Procedia PDF Downloads 416

Authors: Krishna Mohan Bathula, Fatou Bintou Loucoubar, FNU Kaleemunnisa, Christelle Scharff, Mark Anthony De Castro

Abstract:

Voice recognition algorithms such as automatic speech recognition and text-to-speech systems with African languages can play an important role in bridging the digital divide of Artificial Intelligence in Africa, contributing to the establishment of a fully inclusive information society. This paper proposes a Deep Learning model that can classify the user responses as inputs for an interactive voice response system. A dataset with Wolof language words ‘yes’ and ‘no’ is collected as audio recordings. A two stage Data Augmentation approach is adopted for enhancing the dataset size required by the deep neural network. Data preprocessing and feature engineering with Mel-Frequency Cepstral Coefficients are implemented. Convolutional Neural Networks (CNNs) have proven to be very powerful in image classification and are promising for audio processing when sounds are transformed into spectra. For performing voice response classification, the recordings are transformed into sound frequency feature spectra and then applied image classification methodology using a deep CNN model. The inference model of this trained and reusable Wolof voice response recognition system can be integrated with many applications associated with both web and mobile platforms.

Keywords: automatic speech recognition, interactive voice response, voice response recognition, wolof word classification

Procedia PDF Downloads 80

Authors: Rami Ayadi, Mohsen Maraoui, Mounir Zrigui

Abstract:

In this paper, we have presented a brief current state of the art for Arabic text representation and classification methods. We decomposed Arabic Task Classification into four categories. First we describe some algorithms applied to classification on Arabic text. Secondly, we cite all major works when comparing classification algorithms applied on Arabic text, after this, we mention some authors who proposing new classification methods and finally we investigate the impact of preprocessing on Arabic TC.

Keywords: text classification, Arabic, impact of preprocessing, classification algorithms

Procedia PDF Downloads 433