In this paper, an image compression algorithm based on the Orthogonal Ramanujan Sums (ORS) is presented. The proposed algorithm is novel in that it combines ORS-based Discrete wavelet Transform (DWT) with the well-kno...
In this paper, an image compression algorithm based on the Orthogonal Ramanujan Sums (ORS) is presented. The proposed algorithm is novel in that it combines ORS-based Discrete wavelet Transform (DWT) with the well-known JPEG image compression technique. The software implementation of the algorithm is done using Python. The results are compared against the standard JPEG compression technique and it is observed that at similar compression ratios, the proposed algorithm achieves upto 69.88% lower MSE, 24.2% higher PSNR and 27.3% higher SSIM compared to JPEG, which makes it a potential candidate for applications where preserving the quality of the images is important.
Hyper spectral images need more memory to store, process, and transmit due to the large amount of information they carry. One of the most effective way to compress hyper spectral images is to represent them as a three...
详细信息
ISBN:
(纸本)9781665473507
Hyper spectral images need more memory to store, process, and transmit due to the large amount of information they carry. One of the most effective way to compress hyper spectral images is to represent them as a three-dimensional tensor. Tensors are multi-dimensional Structures. Tensors have a wide range of applications in numerical linear algebra, chemometrics, data mining, signalprocessing, statics, data mining, and machine learning. For dimensional reduction of tensors, many tensor decomposition methods have been developed, which we can adapt to Hyper spectral image compression. The proposed method uses Discrete wavelet Transform and Higher-Order Orthogonal Iteration Tucker Decomposition method to compress the Hyper spectral image. The simulation results were compared with 3 more tensor decomposition techniques. Four real hyper spectral images were used in the experimentation: Pavia University (610 X 340 X 103), Indian Pines adjusted (145 x 145 x 200), Salinas image ( 512 x 217 x 224), and Abu-beach (150 X 150 X 102). After processing, the perceptional quality of the Hyper spectral images is evaluated using PSNR and SSIM. When compared to the other three methods, the proposed method offers good PSNR and SSIM with High Compression.
This paper mainly focuses on the evaluation and prediction of sports competition results, and uses the methods of feature extraction, CRITIC weighted analysis and machine learning algorithm to discuss and analyze. Fir...
详细信息
ISBN:
(数字)9798350360240
ISBN:
(纸本)9798350384161
This paper mainly focuses on the evaluation and prediction of sports competition results, and uses the methods of feature extraction, CRITIC weighted analysis and machine learning algorithm to discuss and analyze. First, after data preprocessing, the dimension of column data is reduced by feature extraction and the evaluation index is obtained. These metrics are then weighted using the CRITIC method to obtain a measure of quantified performance. Secondly, the correlation analysis model is introduced to prove the existence of momentum. Finally, by comparing the performance of XGBoost, Extra Trees and Random Forest algorithm in training and prediction, it is found that XGBoost algorithm has the best performance in AUC of test set, reaching 95.05%. The XGBoost parameters were further optimized by Bayesian optimization, and the signal was reconstructed by combining wavelet analysis to improve the accuracy and robustness of the model.
image fusion combines information from multiple images in a single image to increase the information processing capabilities. Infrared and visible image fusion is finding a lot of applications these days, as it brings...
详细信息
ISBN:
(纸本)9781665428644
image fusion combines information from multiple images in a single image to increase the information processing capabilities. Infrared and visible image fusion is finding a lot of applications these days, as it brings complementary information from the source images in one image. The performance of image fusion is affected by noise and artifacts, which tend to creep in during fusion process. Pre-processing of source images is generally done to improve the quality of final fused image. In this paper, a general framework for fusion of infrared and visible image is proposed for investigating role of preprocessing on quality of image fusion. Various arrangements of image enhancement and denoising techniques are used to investigate the role of preprocessing in improving the efficiency of image fusion. The subjective and objective performance evaluations show that joint use of denoising and enhancement improves the performance of infrared and visible image fusion.
Cotton is one of the most vital cash crops cultivated around the globe and its yield directly influences the economy and the livelihood of huge number of people associated with it. The major loss of cotton occurs due ...
详细信息
ISBN:
(数字)9798350361186
ISBN:
(纸本)9798350361193
Cotton is one of the most vital cash crops cultivated around the globe and its yield directly influences the economy and the livelihood of huge number of people associated with it. The major loss of cotton occurs due to damage incurred by microbial pathogens that infect the cotton crop. The course of infection and their effect on the growth and yield of the plant varies depending upon the type of disease. Appropriate treatment for the specific type of disease could prevent the spread of infection and also reduce the use of pesticide that is usually done during the empirical treatment techniques. Identification of the type of cotton plant disease is critical and needs to be done at a faster pace. imageprocessingbased machine learning approach is found to be a better option for this application. This article deals with development of the machine learning model capable of identifying the different types of cotton plant diseases using imageprocessing. Foliar images of diseased cotton plants are collected from public databases and preprocessing operations like resizing, filtering and contrast enhancement are performed followed by k-means clustering for segmentation of diseased part of the leaf. image decomposition is performed using discrete wavelet transform with Daubechies wavelet at five levels and the mean, standard deviation, and entropy of the coefficients are used for building the machine learning model. Classification with Artificial Neural Network trained with backpropagation algorithm has a classification accuracy of 97% indicating the veracity of the model. Hence, the developed model could be used as for developing a user-friendly interface that could be tapped for real-time applications.
Excessive angular decision diffusion-weighted imaging (HARDWI) is an effective approach for visualizing tissue microstructures, which are otherwise hard to look at the usage of conventional MRI technology. To attain t...
详细信息
ISBN:
(数字)9798350370249
ISBN:
(纸本)9798350370270
Excessive angular decision diffusion-weighted imaging (HARDWI) is an effective approach for visualizing tissue microstructures, which are otherwise hard to look at the usage of conventional MRI technology. To attain this, HARDWI relies on sign-processing techniques to reconstruct images from tremendously anisotropic diffusion-weighted alerts. Novel signalprocessing strategies advanced in current years have further advanced the abilities of HARDWI. These strategies leverage the sparsity of diffusion signal in more than one angular guideline, allowing the reconstruction of man or woman diffusion occasions within the specimen’s microstructure. wavelet rework-primarily based strategies allow for the advanced decision of the reconstructed images, extended evaluation, and stepped forward signal-to-noise ratio in the resulting photographs. Other strategies, inclusive of diffusion orientation transform, have been advanced to estimate the orientation of fiber tracts better. Furthermore, spatially-variable regularization techniques have been advanced for advanced estimation of the diffusion parameters at a voxel stage. The combination of these signalprocessing techniques has advanced the application of HARDWI to a spread of biomedical applications, such as neuroanatomy, tumor imaging, and small animal studies.
This review paper explores the fundamental importance of fractional wavelet filter (FrWF) and discrete wavelet transform (DWT) techniques in signalprocessing. These techniques are thoroughly examined and serve as the...
This review paper explores the fundamental importance of fractional wavelet filter (FrWF) and discrete wavelet transform (DWT) techniques in signalprocessing. These techniques are thoroughly examined and serve as the foundation for many different applications, including data compression, telecommunications, and medical diagnostics. DWT’s multi-resolution analysis, facilitated by breaking signals into various scales, plays a vital role in image compression, denoising, and feature extraction. FrWF, designed for memory-constrained environments, notably reduces memory overhead through external memory utilization. The adaptive memory allocation technique enhances memory efficiency, pivotal for real-time applications like sensors and cameras. These techniques, distinct yet impactful, bolster science and technology. This review highlights their vital roles, progressions, and potential in reshaping memory-restricted signalprocessing, redefining paradigms.
Multimodal image fusion is a method of fusing images of different modalities into one image without losing the overall meaning of the input images. The homomorphic method is one method to enhance digital images by inc...
详细信息
Optical images and synthetic aperture radar (SAR) remote sensing images are highly complementary information for Earth observation and applications. Accurate registration of these two data types is essential for subse...
详细信息
ISBN:
(数字)9798331515669
ISBN:
(纸本)9798331515676
Optical images and synthetic aperture radar (SAR) remote sensing images are highly complementary information for Earth observation and applications. Accurate registration of these two data types is essential for subsequent collaboration processing and analysis. However, traditional methods struggle to achieve effective registration due to nonlinear radiometric disparities and geometric distortions. Meanwhile, supervised deep learning approaches require costly manual annotations. To address these issues, this paper proposes an unsupervised registration method based on domain adversarial wavelet learning. We innovatively design an inter-domain adversarial mutual transfer (IAT) module to alleviate cross-modal differences while preserving structural features. Additionally, wavelet transform is employed at different scales for feature matching, to mitigate radiometric disparities and enhance feature recognizability. By incorporating an attention mechanism, the network structure is optimized through spatial alignment and image similarity. The proposed method outperforms other state-of-the-art methods in both quantitative and qualitative aspects of benchmark datasets.
The emergence of deep neural architectures greatly enhanced the accuracy of salient region detection algorithms that plays a vital role in computer vision applications. However, the accurate extraction of regions with...
详细信息
ISBN:
(纸本)9781728176055
The emergence of deep neural architectures greatly enhanced the accuracy of salient region detection algorithms that plays a vital role in computer vision applications. However, the accurate extraction of regions with fine boundaries still remains as a challenge. In this work, an attention based wavelet Convolutional Neural Network (WCNN) is implemented that efficiently extracts the spatial, spectral and semantic features of the image in multiple resolution and it turns out to be suitable for locating the visually salient regions. Further enhancement of the fine boundaries of the predicted map is made possible by the inclusion of a combinational loss function of balanced cross entropy loss, SSIM loss and edge loss. The effectiveness of the method is evaluated using three benchmark datasets and the results shows better performance achieving a minimum Mean Absolute Error (MAE) of 0.032 and maximum F-measure of 0.938.
暂无评论