International Journal of Image, Graphics and Signal Processing @ijigsp
Статьи журнала - International Journal of Image, Graphics and Signal Processing
Все статьи: 1128
Статья научная
Speech is the natural mode of communication between humans. Human-to-machine interaction is gaining importance in the past few decades which demands the machine to be able to analyze, respond and perform tasks at the same speed as performed by human. This task is achieved by Automatic Speech Recognition (ASR) system which is typically a speech-to-text converter. In order to recognize the areas of further research in ASR, one must be aware of the current approaches, challenges faced by each and issues that needs to be addressed. Therefore, in this paper human speech production mechanism is discussed. The various speech recognition techniques and models are addressed in detail. The performance parameters that measure the accuracy of the system in recognizing the speech signal are described.
Бесплатно
An improved image compression algorithm using wavelet and fractional cosine transforms
Статья научная
The most significant parameters of image processing are image resolution and speed of processing. Compressing the multimedia datasets, which are rich in quality and volume is challenging. Wavelet based image compression techniques are the best tools for lossless image compression, however, they suffer by low compression ratio. Conversely fractional cosine transform based compression is a lossy compression technique with less image quality. In this paper, an improved compression technique is proposed by using wavelet transform and discrete fractional cosine transform to achieve high quality of reconstruction of an image at high compression rate. The algorithm uses wavelet transform to decompose image into frequency spectrum with low and high frequency sub bands. Application of quantization process for both sub bands at two levels increases the number of zeroes, however rich zeroes from high frequency sub bands are eliminated by creating the blocks and then storing only non-zero values and kill all blocks with zero values to form reduced array. The arithmetic coding method is used to encode the sub bands. The Experimental results of proposed method are compared with its primitive two dimensional fractional cosine and fractional Fourier compression algorithms and some significant improvements can be observed in peak signal to noise ratio and self-similarity index mode at high compression ratio.
Бесплатно
An optimized architecture of image classification using convolutional neural network
Статья научная
The convolutional neural network (CNN) is the type of deep neural networks which has been widely used in visual recognition. Over the years, CNN has gained lots of attention due to its high capability to appropriately classifying the images and feature learning. However, there are many factors such as the number of layers and their depth, number of features map, kernel size, batch size, etc. They must be analyzed to determine how they influence the performance of network. In this paper, the performance evaluation of CNN is conducted by designing a simple architecture for image classification. We evaluated the performance of our proposed network on the most famous image repository name CIFAR-10 used for the detection and classification task. The experiment results show that the proposed network yields the best classification accuracy as compared to existing techniques. Besides, this paper will help the researchers to better understand the CNN models for a variety of image classification task. Moreover, this paper provides a brief introduction to CNN, their applications in image processing, and discuss recent advances in region-based CNN for the past few years.
Бесплатно
Analog Document Search Using CRNN and Keyphrase Extraction
Статья научная
There seems to be a peculiar trend in the way information is now used, moving to digital media not just for the newspapers but for books as well. With advances in Optical Character Recognition (OCR), Style Transfer Mapping (STM), and efficient key phrasing, we are now able to digitalize the document to a form that can be read across multiple platforms and searched efficiently. It provides users with the ease of searching for relevant documents without the tedious process of manual searching. We propose a system that uses the CRNN model to detect English characters in the document with high accuracy. We then pair it with a hybrid keyphrasing technique, which uses Positional Rank as its Graph based rank and re-rank the key phrases using the C-Value method. This process allows us to automatically digitize the printed document and summarise it to provide high-quality keyphrases, which can be used to efficiently search and retrieve relevant printed documents.
Бесплатно
Analysis and Estimation of Noise in Embedded Medical Images
Статья научная
Patient information is embedded inside the medical images for the storage or transmission, or healthcare applications. In medical image processing, various types of noises corrupt the image quality. There is a need of measure specific noise for a particular image is required for the evaluation of robustness for embedding techniques used for hiding patient information in medical images. It is very important to obtain precise images to facilitate accurate analysis and estimation of noise in embedded medical image. The current work is focused towards studying the effect of specific noise which affect particular medical image. The strength of the medical image is tested by introducing several attacks to the embedded medical images. The statistical quantity measures like peak signal-to-noise ratio (PSNR), signal-to-noise ratio (SNR) and normalized root mean square error (NRMSE) are employed to measure the quality of the output medical image.
Бесплатно
Analysis and detection of content based video retrieval
Статья научная
Content Based Video Retrieval (CBVR) System has been investigated over past decade it’s rooted in many applications like developments and technologies. The demand for extraction of high level semantics contents as well as handling of low level contents in video retrieval systems are still in need. Hence it motivates and encourages many researchers to discover their knowledge across CBVR domain and contribute their work to make the system more effective and useful in developing the system application. This paper highlights comprehensive and extensive review of CBVR techniques for detection of region of interest in a given video. The experiment is carried out for the detection of ROI using ACF detector. The detection rate of ROI is observed competitive and satisfactory.
Бесплатно
Analysis of Abdominal ECG Signal for Fetal Heart Rate Estimation Using Adaptive Filtering Technique
Статья научная
This paper presents a method for fetal heart rate estimation from an abdominal electrocardiogram (ECG) signal based on adaptive filter analysis using least mean square (LMS) adaptive filtering algorithm in order to determine the health status of a baby in its mother's womb. The fetal ECG signal is extracted from abdominal ECG containing other sources of interference using the maternal ECG signal obtained from mother's chest cavity as the reference signal. Interference/noise model used for this work include the power-line noise, the white noise and the unwanted propagating maternal ECG signal. Thereafter, the heart rate is estimated using an automated peak voltage measurement algorithm at 75 percent threshold voltage. It is found that irrespective of the estimated heart rate of the baby, 100 percent estimation is achieved at signal-to-noise ratio (SNR) greater than or equal to -31dB.
Бесплатно
Analysis of Arithmetic and Huffman Compression Techniques by Using DWT-DCT
Статья научная
In the recent era, digital contents are exchanging over the internet and it has increased exponentially. Sometimes, we need small sizes to share the real world, because of narrow bandwidth. Hence, the data compression concept came in limelight to utilize the storage capacity and available bandwidth efficiently. This paper presents an analysis of Arithmetic and Huffman compression techniques based on a hybrid combination of the DWT-DCT techniques. The input image is decomposed up to the 3rd level by using the DWT and then Arithmetic & Huffman coding is applied separately on quantized sub-bands on 2nd as well as 3rd level coefficients from approximation sub-bands to get a high compression ratio and high peak signal-to-noise ratio values. On the third level approximation sub-band, the DCT method is applied to reduce the blocking effect. Simulation results show that the Arithmetic coding exhibits higher CR than Huffman coding, but smaller PSNR values.
Бесплатно
Статья научная
Scaling behavior is an indicator of the lack of characteristic time scale, and the existence of long-range correlations related to physiological constancy preservation. To investigate the fluctuations of the sleep electroencephalogram (EEG) over various time scales during different sleep stages detrended fluctuation analysis (DFA) is studied. The sleep EEG signals for analysis were obtained from the Sleep-EDF Database available online at the PhysioBank. The DFA computations were performed in different sleep stages. The scaling behavior of these time series was investigated with detrended fluctuation analysis (window size: 50 to 500). The results show that the mean values of scaling exponents were lower in subjects during stage 4 and standard deviation of scaling exponents of stage 4 was larger than that of the other stages. In contrast, the mean value of scaling exponents of stage 2 was larger, while a small variation of scaling exponent is observed at this stage. Therefore, DFA has a more stable behavior in stage 2, whereas the random variability and unpredictable behavior of DFA can be observed in the stage 4. In conclusion, scaling exponent indices are efficacious in quantifying EEG signals in different sleep stages.
Бесплатно
Статья научная
Neurophysiological parameters revealed by resting-state electroencephalography (rsEEG) may be helpful in the diagnosis of various brain diseases like Epilepsy, Alzheimer’s, depressive disorders, and many others. Due to the abrupt onset of seizures, Epilepsy is a chronic nerve illness that interferes with an epileptic patient's regular everyday activities. However, manual investigation of EEG for finding epileptiform discharges by skilled neurologists is a laborious, time-consuming, and error-prone process. It might cause a significant delay in providing clinical care to a person who could have epilepsy. This work offers a straightforward method for analyzing EEG data for the purpose of identifying epileptic features by iteratively simulating multiple deep learning models. It also attempts to include big data analytics for handling the challenge of analyzing the mountain of unstructured EEG data, available and accessible in numerous formats. In contrast to the state-of-the-art works, the performance scores of the proposed methods show significant improvement for the considered assessment parameters. Additionally, after testifying the performance of this proposed technique for relevant datasets, its application can be extended to identify other neurodegenerative disorders as well. Therefore, this study can assist physicians and healthcare professionals in the efficient care and treatment of patients with mental health issues.
Бесплатно
Статья научная
Presently considerable amount of work has been done in tele-monitoring which involves the transmission of bio-signals and medical images in the wireless media. Intelligent exchange of bio-signals amongst hospitals needs efficient and reliable transmission. Watermarking adds “ownership” information in multimedia contents to prove the authenticity, to verify signal integrity, or to achieve control over the copy process. This paper proposes a novel session based blind watermarking method with secret key by embedding binary watermark image into (Electrocardiogram) ECG signal. The ECG signal is a sensitive diagnostic tool that is used to detect various cardio-vascular diseases by measuring and recording the electrical activity of the heart in exquisite detail. The first part of this paper proposes a multi-resolution wavelet transform based system for detection ‘P’,‘Q’,‘R’,‘S’,‘T’ peaks complex from original ECG signal of human being. ‘R-R’ time lapse is an important component of the ECG signal that corresponds to the heartbeat of the concerned person. Abrupt increase in height of the ‘R’ wave or changes in the measurement of the ‘R-R’ interval denote various disorders of human heart. Similarly ‘P-P’, ‘Q-Q’, ‘S-S’, ‘T-T’ intervals also correspond to different disorders of heart and their peak amplitude envisages other cardiac diseases. In this proposed method the ‘P Q R S T’-peaks are marked and stored over the entire signal and the time interval between two consecutive ‘R’-peaks and other peaks interval are measured to detect anomalies in behavior of heart, if any. The peaks are achieved by the composition of Daubechies sub-bands wavelet of original ECG signal. The accuracy of the P, QRS and T components detection and interval measurement is achieved with high accuracy by processing and thresholding the original ECG signal. The second part of the paper proposes a Discrete Wavelet Transformation (DWT) and Spread Spectrum based watermarking technique. In this approach, the generated watermarked signal having an acceptable level of imperceptibility and distortion is compared to the Original ECG signal. Finally, a comparative study is done for the intervals of two consecutive ‘R-R’ peaks, ‘P-R’, ‘Q-T’, ‘QTc’, QRS duration, cardiac output between original P, QRS and T components detected ECG signal and the watermarked P,QRS and T components detected ECG signal.
Бесплатно
Analysis of Rotman Lens Antenna for Different Substrates for Circular Contour
Статья научная
This paper presents a trifocal Rotman lens design approach. The effect due to change of substrate on the circular contour is observed. The shape of the beam contour is taken as circular. Different substrates can be used for the fabrication of the lens. Three different materials have been used to fabricate the lens antenna .A three beam prototype feeding five element antenna array working in ISM band has been simulated using RLD1.7.Effects on the performance of the antenna is observed.
Бесплатно
Статья научная
A certain number of researchers have utilized uni-modal bio-metric traits for gender classification. It has many limitations which can be mitigated with inclusion of multiple sources of biometric information to identify or classify user’s information. Intuitively multimodal systems are more reliable and viable solution as multiple independent characteristics of modalities are fused together. The objective of this work is inferring the gender by combining different biometric traits like face, iris, and fingerprints of same subject. In the proposed work, feature level fusion is considered to obtain robustness in gender determination; and an accuracy of 99.8% was achieved on homologous multimodal biometric database SDUMLA-HMT (Group of Machine Learning and Applications, Shandong University). The results demonstrate that the feature level fusion of Multimodal Biometric system greatly improves the performance of gender classification and our approach outperforms the state-of-the-art techniques noticed in the literature.
Бесплатно
Analysis of the Error Pattern of HMM based Bangla ASR
Статья научная
Speech Recognition research has been ongoing for more than 80 years. Various attempts have been made to develop and improve speech recognition process around the world. Research on ASR by machine has attracted much attention over the last few decades. Bengali is largely spoken all over the world. There are lots of scopes yet to explore in the research regarding offline automatic Bangla speech recognition system. In our work, a moderate size speech corpus and a HMM based speech recognizer have been built to analyze the error pattern. Audio recordings have been collected from different persons in both quiet and noisy area. Live test has been carried out also to check the performance of the model individually. The percentage of the error and the percentage of correction with the created models are presented in this paper along with the results obtained during the live test. Finally, the results are analyzed to get the error pattern needed for future development.
Бесплатно
Статья научная
Clinically, ECG is a potential tool used in the applications evaluating electrical activities of heart, functioning and providing solution to problems associated with it. Among such, the relationship (correlation) between respiration and electrocardiographic signal has attracted attention in past decades. In this research, a Welch spectrum estimation approach was utilized for normalizing cross spectrum analysis between these two signals. This approach can be useful while diagnosing diseases like pulmonary embolism, coronary lung diseases, Deep vein thrombosis and other diseases related to heart, from the knowledge of existing coherence bonding between these signals. This research applies the above approach to human subjects, whose ECG and respiratory signal annotations has been evaluated and were sampled at 100 samples/ second (sampling rate). The different respiratory signals are taken from Chest (CRSP), abdominal (ARSP) and oronasal regions (NRSP). The annotated signals for all the four subjects, discussed in this paper were obtained through a non-invasive test, which is medically well known as impedance phlebography, or impedance plethysmography. The numbers of samples, under the analysis were 6000 for each signal. The data was acquired from recording database of physionet. For this examination the mean square coherence (MSC) was chosen as an excellent candidate. The results imply that the mean of MSCs is found continuously decreasing in chest respiration. Secondly, the results showed maximum coherence between ECG and corresponding respiratory signal in three subjects is in Abdominal (ARSP) region (i.e. having maximum value greater than 0.5). Lastly, above analysis was analyzed over the fourth subject's data and under observation it was found, exceptionally that, the value of coherence for all respiratory patterns showed a poor functional association or simply coherence between the signal i.e.Coh2 below 0.5 in the abdominal region (ref.Fig.5) and the reason suggested could be chronic lung disease while the results show higher values, that is between (0.5 < coherence <1) in other two. Further, we show that the coherence peak reflects that the one physiological signal is synchronized with another signal of same nature at a particular frequency, here it is 0 to 35 Hz frequency band and combined analysis is shown through a Boxplot, from three regions showing maximum value of coherence upper quartile in abdominal region for three healthy subjects with maximum value of peak in the same region. This paper also presents a platform to dissolve the problem pertaining in an individual related to deep vein thrombosis, hypoxemia (blood level <90%) [16] and related diseases by estimating the coupling associated between saturated oxygen content (SO2) with respiratory patterns, in order to detect dysfunctioning clinically, also for efficient heart working. Thus, the research shows successful attempt to investigate the interaction of the PS of ECG signal and respiratory signals. The work presented in this paper can further be extended by adopting different method and either by defining a vector array element for maximum number of coherence value that could be beneficial for detecting diseases like sleep apnea on basis of minimum or maximum occurrence of peaks.
Бесплатно
Analysis on skin colour model using adaptive threshold values for hand segmentation
Статья научная
The hand gesture recognition system is the hottest topic for the human-machine interaction and computer vision fields. The hand gesture recognition system is still a challenging research area in computer vision for human-computer interaction because of various device conditions, various illumination effects, and very complex background. The recognition of hand gestures used in various application areas: such as sign language recognition, man-machine interaction, human-robot interaction, and intelligent device control and many other application areas. The robust detection of hand in hand gesture recognition system has become a challenging task due to clutter background, dynamic background, and various illumination conditions in real-world conditions. Segmentation is the partioning/separating the foreground hand region from the background region in an image. Segmentation is also pre-processing steps of the hand gesture recognition system. The recognition accuracy will increase if the hand region correctly detected. So, hand region detection is the main important step for the hand gesture recognition system.
Бесплатно
Статья научная
A new local-topological approach to describe the spatial and angular distributions of polarization parameters of multiply scattered optically anisotropic biological layers of laser fields is considered. A new analytical parameter to describe the local polarization structure of a set of points of coherent object fields, the degree of local depolarization (DLD), is introduced for the first time. The experimental scheme and the technique of measuring coordinate distributions (maps) of DLD The new method of local polarimetry was experimentally tested on histological specimens of biopsy sections of operatively extracted breast tumors. The measured DLD maps were processed using statistical, autocorrelation and scale-sampling approaches. Markers for differential diagnosis of benign (fibroadenoma) and malignant (sarcoma) breast tumors were defined.
Бесплатно
Analyzing defects of solar panels under natural atmospheric conditions with thermal image processing
Статья научная
Sun is an ultimate source of energy, which is clean, inexhaustible and safe for environment. The energy obtained from sun is known as solar energy. When solar radiations fall on earth surface solar cells convert these solar radiations into electrical energy. Solar cells are one of the important components of solar panels. Many solar cells combine in series or in parallel to form solar module and solar panels. A solar photovoltaic array is a combination of solar panels and is installed in open atmospheric condition. The natural conditions such as dust and dirt, shade of tree affect operation of solar panels. These natural conditions cannot be avoided but can be analyzed using visual inspection and thermal camera. The visual inspection approach is useful only when the defects are visible by naked human eye but when defects cannot be visualized by naked human eye thermo-graphical approach is used. This paper discusses the identification of various defects in solar panels by applying image processing technique applied for thermal images under natural atmospheric conditions for visual inspection, shading effect of tree, dust and dirt deposition effects on solar panels using thermal imaging camera.
Бесплатно
Anti-Forensics of JPEG Images using Interpolation
Статья научная
The quantization artifacts and blocking artifacts are the two significant properties for identifying the forgery in a JPEG compressed image. There are some techniques for JPEG compressed images that can remove these artifacts resulting no traces for forgery. These methods are referred as anti-forensic methods. A forger may perform some post-operations to disturb the underlying statistics of JPEG images to fool current forensic techniques. These methods create noise and reduce the image quality. In this paper we apply three different interpolation techniques namely nearest neighbor, bilinear and bicubic techniques to remove JPEG artifacts. The experimental results show that the bicubic interpolated images are found to be of better quality as compare to the nearest neighbor and bilinear interpolated images with no JPEG artifacts. For quality analysis of these interpolation methods on the images three popular quality metric are used. The proposed method is very simple to perform. This interpolation based method is applicable to both single and double JPEG compression.
Бесплатно
Статья научная
Medical experts often examine hundreds of x-ray images searching for salient features that are used to detect pathological abnormalities. Inspired by our understanding of the human visual system, automated salient features detection methods have drawn much attention in the medical imaging research community. However, despite the efforts, detecting robust and stable salient features in medical images continues to constitute a challenging task. This is mainly attributed to the complexity of the anatomical structures of interest which usually undergo a wide range of rigid and non-rigid variations. In this paper, we present a novel appearance-based salient feature extraction and matching method based on sparse Contourlet-based representation. The multi-scale and directional capabilities of the Contourlets is utilized to extract salient points that are robust to noise, rigid and non-rigid deformations. Moreover, we also include prior knowledge about local appearance of the salient points of the structure of interest. This allows for extraction of robust stable salient points that are most relevant to the anatomical structure of interest.
Бесплатно
