Papers by Kamrul Hasan Talukder
2D image compression using concurrent wavelet transform
International Conference on Graphic and Image Processing (ICGIP 2011), Oct 1, 2011
ABSTRACT In the recent years wavelet transform (WT) has been widely used for image compression. A... more ABSTRACT In the recent years wavelet transform (WT) has been widely used for image compression. As WT is a sequential process, much time is required to transform data. Here a new approach has been presented where the transformation process is executed concurrently. As a result the procedure runs first and the time of transformation is reduced. Multiple threads are used for row and column transformation and the communication among threads has been managed effectively. Thus, the transformation time has been reduced significantly. The proposed system provides better compression ratio and PSNR value with lower time complexity.
Message Sequence Charts (MSCs) can be used effectively to specify the communicating threads in th... more Message Sequence Charts (MSCs) can be used effectively to specify the communicating threads in the way where high-level transition systems is used to capture the control flow of the system components. This specification is amenable to formal verification. Here, we present the way how we can specify the communicating threads using MSCs and how we can verify the correctness of the concurrent wavelet transform used for the purpose of image compression. We have used Symbolic Model Verifier (SMV) for the use of formal verification. The introduction of wavelet transform for image compression followed by the MSC based specification for concurrent wavelet transform is presented. The concurrency in the threads for concurrent wavelet transform along with the grammar of the syntax formulated is accessed. At the end, the verification result of the concurrent wavelet transform using the SMV program is presented.
We have proposed a wavelet-based image transmission scheme in this paper which has used discrete ... more We have proposed a wavelet-based image transmission scheme in this paper which has used discrete wavelet transform to transform a digital image from spatial domain into frequency domain. Haar wavelet transformation has been chosen and used for wavelet transformation phase. Use of concurrent computing has significantly reduced computation time overhead as well as the transmission time up to a great extent as compared to the existing sequential discrete wavelet based transmission. The experimental result shows that the proposed scheme maintains the accuracy of reconstructed image.
Multi-threaded concurrent wavelet based image compression and image editing
Since the starting of the use of wavelet in the 1980s as an alternative to the windowed fourier t... more Since the starting of the use of wavelet in the 1980s as an alternative to the windowed fourier transform for signal analysis, it emerged as a cutting edge technology within the field of image analysis and in computer graphics. However, it is a very computationally intensive process which requires innovative and computationally efficient method. The concurrent wavelet transform might be a solution to it. Here multiple threads are engaged to transform the image data concurrently. This paper uses the Haar wavelet technique. The compression method has been evaluated in respect of compression ratio and PSNR. Additionally, the perception based subjective (Mean Opinion Score (MOS)) and objective (Picture Quality Scale (PQS)) evaluation methods are also applied and their effectiveness is evaluated. The multi-resolution nature of the Haar wavelets facilitates the easy inclusion of image editing in the transformed images. This editing method is also presented with better result.
World Academy of Science, Engineering and Technology, International Journal of Computer, Electrical, Automation, Control and Information Engineering, Nov 27, 2009
The scientific community has invested a great deal of effort in the fields of discrete wavelet tr... more The scientific community has invested a great deal of effort in the fields of discrete wavelet transform in the last few decades. Discrete wavelet transform (DWT) associated with the vector quantization has been proved to be a very useful tool for the compression of image. However, the DWT is very computationally intensive process requiring innovative and computationally efficient method to obtain the image compression. The concurrent transformation of the image can be an important solution to this problem. This paper proposes a model of concurrent DWT for image compression. Additionally, the formal verification of the model has also been performed. Here the Symbolic Model Verifier (SMV) has been used as the formal verification tool. The system has been modeled in SMV and some properties have been verified formally.
With the increasing growth of technology and the entrance into the digital age, we have to handle... more With the increasing growth of technology and the entrance into the digital age, we have to handle a vast amount of information every time which often presents difficulties. So, the digital information must be stored and retrieved in an efficient and effective manner, in order for it to be put to practical use. Wavelets provide a mathematical way of encoding information in such a way that it is layered according to level of detail. This layering facilitates approximations at various intermediate stages. These approximations can be stored using a lot less space than the original data. Here a low complex 2D image compression method using wavelets as the basis functions and the approach to measure the quality of the compressed image are presented. The particular wavelet chosen and used here is the simplest wavelet form namely the Haar Wavelet. The 2D discrete wavelet transform (DWT) has been applied and the detail matrices from the information matrix of the image have been estimated. The reconstructed image is synthesized using the estimated detail matrices and information matrix provided by the Wavelet transform. The quality of the compressed images has been evaluated using some factors like Compression Ratio (CR), Peak Signal to Noise Ratio (PSNR), Mean Opinion Score (MOS), Picture Quality Scale (PQS) etc.
An Efficient Hashing Method for Exact String Matching Problems
Springer eBooks, 2022
Communication Among Multiple Threads in the Concurrent Wavelet Transformation for Image Compression
ABSTRACT With the increasing growth of technology and the entrance into the digital age, a huge a... more ABSTRACT With the increasing growth of technology and the entrance into the digital age, a huge amount of image data needs to have handled every time which often presents difficulties. So, these data must be stored and retrieved in an efficient and effective manner, in order for it to be put to practical use. In this connection, wavelet transform is used for decades as it emerged as a cutting edge technology within the field of image analysis and in computer graphics. Due to the intensive computation of wavelet transform of the image data, the multi-threaded concurrent computing is considered to be a solution. The Message Sequence Charts (MSCs) are used effectively to specify the communicating threads in the way where high-level transition systems are used to capture the control flow of the system components and MSCs to describe the non-atomic component interactions. This specification is efficiently amenable to formal verification. We present the way how to specify the communicating threads using MSCs. The compression method and the use of wavelet transform for compression are presented. Moreover, the MSC based specification along with the modified concurrency in the threads for concurrent wavelet transform are also presented.
Sentiment Analysis of Bengali Text using CountVectorizer with Logistic Regression
Sentiment analysis refers to find the polarity of a text. Sentiment analysis or opinion mining on... more Sentiment analysis refers to find the polarity of a text. Sentiment analysis or opinion mining on social media data is an emerging topic in research. In Bangladesh, there remains a few researches on Bengali text. In Natural Language Processing, text document can be represented by vector which is a very efficient way to find the similarities among the text. Word2vec is a technique to represent each word as a vector and Doc2vec represent a whole sentence or document as a vector. CountVectorizer is another technique to transform a corpora of text to a vector form along with the token counts. There remains a few researches using Word2vec and Doc2vec. And recent studies showed that CountVectorizer outperforms both of the Doc2vec and Word2vec techniques. In this study, we have used a dataset of 7,000 Bengali text to apply sentiment classification into two classes: positive and negative. We have extracted the feature using CountVectorizer and then have applied the machine learning method: Logistic Regression for the classification task. The result shows higher accuracy than the previous works.
Lung Cancer Detection from Histopathological Images Using Deep Learning
Analysis of Hand Movement from Surface EMG Signals Using Artificial Neural Network
Khulna University studies, Nov 24, 2022
The recognition of Bangla handwritten numerals (BHNR) has recently emerged as a very interesting ... more The recognition of Bangla handwritten numerals (BHNR) has recently emerged as a very interesting area for machine learning and pattern recognition research. Recently, the technology for character or object recognition has also advanced. Bangla handwritten number recognition can serve as a foundation for creating an Optical Character Recognition (OCR) in the Bangla language. However, the lack of a sizable and accurate dataset makes Bangla's handwritten numeral recognition study insufficient in contrast to that of other well-known languages. Similar to MNIST for English digits, NumtaDB is by far the largest dataset collection for handwritten digits in the Bangla language. The most used datasets for the recognition of Bangla handwritten numerals in the past were NumtaDB, CMATERdb, and ISI. The majority of approaches now in use rely on feature extraction and a few outdated machine learning algorithms. Although some approaches operate quickly enough to meet practical demands, they are not always accurate. Even while certain techniques work quite well for languages other than Bangla, they still require improvement. Convolutional Neural Networks (CNN), in particular, are demonstrating excellent achievements in this discipline with higher accuracy. In this work, we've used custom CNN architectures to build our model to recognize digits using all the existing datasets with a high degree of accuracy. Our CNN model shows an average of 98% accuracy recognizing Bangla numeric in respect of above datasets. We have cross verified our model with mixed datasets and the result is also promising.
Smart Environment Index Prediction of Smart City Using Polynomial Regression
Khulna University studies, Nov 17, 2022
Khulna University studies, Nov 23, 2022
The recent world is facing a new pandemic which is caused by a virus named Coronavirus. Its fast ... more The recent world is facing a new pandemic which is caused by a virus named Coronavirus. Its fast mutation capability makes the situation worse affecting all the countries. Handling the virus is a challenging task now as there is still no permanent remedy for this. The doctors, engineers, scientists all are working together to fight against the virus. Revealing the genome sequencing and total structure of the virus paves the way for more research on this topic. Many researchers and scientists are working relentlessly on mutation analysis. Since spike proteins are one of the most important parts of SARS-CoV-2 for affecting humans, scientists are working for vaccine and drug discovery targeting S protein. Many Machine learning, Artificial Intelligence, Deep Learning methods are used on the genome datasets to detect the mutation position and predict further insights. The goal of this work is to predict the most probable next-generation Spike Protein sequence of SARS-CoV-2. We have proposed a model that uses the Encoder-Decoder based LSTM model on date-wise ordered protein sequence data of S-protein. This has worked effectively on predicting next generation sequence of S protein. We compared our model with other deep learning models i.e. CNN-LSTM and Attention-based LSTM. We also experimented our model with large datasets as well as with small datasets, and the results of the tests are effective and efficient in both ways.
arXiv (Cornell University), Sep 26, 2010
Message Sequence Charts (MSCs) are an appealing visual formalism mainly used in the early stages ... more Message Sequence Charts (MSCs) are an appealing visual formalism mainly used in the early stages of system design to capture the system requirements. However, if we move towards an implementation, an executable specifications related in some fashion to the MSC-based requirements must be obtained. The MSCs can be used effectively to specify the bus protocol in the way where high-level transition systems is used to capture the control flow of the system components of the protocol and MSCs to describe the non-atomic component interactions. This system of specification is amenable to formal verification. In this paper, we present the way how we can specify the bus protocols using MSCs and how these specifications can be translated into program of verification tool (we have used Symbolic Model Verifier (SMV)) for the use of formal verification. We have contributed to the following tasks in this respect. Firstly, the way to specify the protocol using MSC has been presented. Secondly, a translator that translates the specifications (described in a textual input file) into SMV programs has been constructed. Finally, we have presented the verification result of the AMBA bus protocol using the SMV program found through the translation process. The SMV program found through the translation process can be used in order to automatically verify various properties of any bus protocol specified.
Performance analysis of back-propagation learning algorithm for handwritten digit classification
This paper presents the investigation of using Multi Layer Perceptions (MLP) for Handwritten Digi... more This paper presents the investigation of using Multi Layer Perceptions (MLP) for Handwritten Digit Classification. The experiment is conducted for an offline digit recognition system. All samples are collected from scanned image files, containing the handwritten digits from different ...
Enhancement of Discrete Wavelet Transform (DWT) for Image Transmission over Internet
Abstract In this paper, a wavelet-based image transmission scheme has been proposed. The proposed... more Abstract In this paper, a wavelet-based image transmission scheme has been proposed. The proposed scheme has used discrete wavelet transform to transform a digital image from spatial domain into frequency domain. For wavelet transformation phase, Haar wavelet ...
Engineering Letters, 2008
Wavelet Transform (WT) has been proved to be a very useful tool for image processing in the recen... more Wavelet Transform (WT) has been proved to be a very useful tool for image processing in the recent years. However, WT is very computationally rigorous process requiring novel and computationally competent method to achieve image compression. The concurrent transformation of the image might be one of the solutions to this problem. In this paper, we investigate and analyze the concurrency in communicating threads that perform the wavelet transformation concurrently. The grammar and its syntax are formulated to specify the communication among threads. Moreover, the system modeled in Symbolic Model Verifier (SMV) is formally verified.
Modeling the Communications in the Multi-threaded System for Concurrent Wavelet Transform to Compress the Image
Abstract Wavelet transform has emerged as a cutting edge technology within the field of image ana... more Abstract Wavelet transform has emerged as a cutting edge technology within the field of image analysis and in computer graphics. However, it is a very computationally intensive process which requires innovative and computationally efficient method. The concurrent ...
Blindness Risk Prediction caused by Diabetic Retinopathy from Retinal Image
Uploads
Papers by Kamrul Hasan Talukder