Objective assessment of the image quality is of keen importance in numerous image processing applications. Various objective quality assessment indexes have been developed for this purpose, of which Peak signal-to-noise ratio (PSNR) is one of the simplest and commonly used. However, it sometimes fails to give result similar to as perceived by the Human Visual System (HVS). This paper presents a modified PSNR metric which is based on HVS characteristics and correlates well with the perceived image quality. It takes into account the error sensitivity, structural distortion and edge distortion in the image. The proposed metric uses RGB model for color images and empirically combine the effects of above distortion types on each of the color plane. Simulation results illustrate the precision and efficiency of the proposed metric in assessing the quality of color images for different types of degradations and show better correlation with the known characteristics of HVS in comparison to conventional PSNR metric.

Automatic human activity recognition (HAR) through computing devices is a challenging research topic in the domain of computer vision. It has widespread applications in various fields such as sports, healthcare, criminal investigation and so on. With the advent of smart devices like smartphones, availability of inertial sensors like accelerometer and gyroscope can easily be used to track our daily physical movements. State-of-the-art deep neural network models like Convolutional Neural Network (CNN) do not need any additional feature extraction for such applications. However, it requires huge amount of data for training which is time consuming, and requires ample resource. Another limiting factor of CNN is that it considers only the features of an individual sample for learning without considering any structural information among the samples. To address the aforesaid issues, we propose an end-to-end fast Graph Neural Network (GNN) which not only captures the individual sample information efficiently but also the relationship with other samples in the form of an undirected graph structure. To the best of our knowledge, this is the first work where the time series data are transformed into a structural representation of graph for the purpose of HAR using sensor data. Proposed model has been evaluated on 6 publicly available datasets, and it achieves nearly 100% recognition accuracy for all the 6 datasets. Source code of this work is available at https://github.com/riktimmondal/HAR-Sensor.

Electroencephalogram (EEG) remains the most immediate, simple, and rich source of information for understanding phenomena related to brain electrical activities. It is certainly a source of basic and interesting information to be extracted using specific and appropriate techniques. The most important aspect in processing EEG signals is to use less co-lateral assets and instrumentation in order to carried out a possible diagnosis; this is the approach of early diagnosis. Advanced estimate spectral analysis can reveal new information encompassed in EEG signals by means of specific parameters or indices. The research proposes a multidimensional approach with a combined use of decimated signal diagonalization (DSD) as basis from which it is possible to work by finding appropriate signal windows for revealing expected information and overcoming signal processing limitations encountered in quantitative EEG. Important information, about the state of the patient under observation, must be extracted from calculated DSD bispectrum. For this aim, it is useful to define an assessment index about the dynamic process associated with the analyzed signal. This information is measured by means of entropy, since the degree of order/disorder of the recorded EEG signal will be reflected in the obtained DSD bispectrum. The general advantage of multidimensional approach is to reveal eventual stealth frequencies “in space and in time” giving a topological vision to be correlated to physical areas which these frequencies emerge from. Long term and sleeping EEG recorded are analyzed, and the results obtained are of interest for an accurate diagnosis of the patient's clinical condition.

Background: Plant diseases significantly affect the crop, so their identification is very important. Correct identification of these diseases is crucial for establishing a good disease control strategy to avoid time and financial losses. In general, machines can greatly reduce the possibility of human error. In particular, computer vision techniques developed through deep learning have paved a way to detect and diagnose these plant diseases on the leaf. Methods: In this work, the model AFD-Net was developed to detect and identify various leaf diseases in apple trees. The dataset is from Kaggle 2020 and 2021 and was financially supported by the Cornell Initiative for Digital Agriculture. An AFD-Net was proposed for leaf disease classification in apple trees and the results of the efficiency of the model are compared with other state-of-the-art deep learning approaches. Results: The results of the experiments in the validation dataset show that the proposed AFD-Net model achieves the highest values of 98.7% accuracy for Plant Pathology 2020 and 92.6% for Plant Pathology 2021 compared to other deep learning models in the original and extended datasets. Discussion: The results also indicate the efficiency of the proposed model in identifying leaf diseases on apple trees for major and minor classes, i.e., for multiple classification.

Fusion of various images aids the rejuvenation of complementary attributes of the images. Similarly, medical image fusion constructs a composite image comprehending significant traits from multimodal source images. Current work exhibits medical image fusion utilizing Laplacian Pyramid (LP) employing DCT. LP decomposes the source medical images as different low pass filtered images, resembling a pyramidal structure. As the pyramidal level of decomposition increases, the quality of the fused image also increases. The proposed technique provides a fused image with better edges and information content from human visual system (HVS) point of view. Qualitative and quantitative analysis of the proposed technique is found to be superior than that of Daubechies complex wavelet transform (DCxWT).

This paper presents a non-linear framework employing a robust polynomial filter for accomplishing enhancement of mammographic abnormalities outcoming from biomedical instrumentation, i.e., X-rays instrumentation. The approach proposed in this paper uses a linear combination of Type-0 and Type-II polynomial filters as a generalized filtering solution to achieve enhancement of mammographic masses and calcifications irrespective of the nature of background tissues. A Type-0 filter provides contrast enhancement, suppressing the ill-effects of background noise. On the other hand, Type-II filter performs edge enhancement leading to preservation of finer details. Contrast improvement index is used as a performance measure to quantify the degree of improvement in contrast of the region-of interest. In addition, estimation of signal-to-noise ratio (in terms of PSNR and ASNR) is carried out to account for the suppression in background noise levels and over-enhancements of the processed mammograms. These measures are used as a mechanism to optimally select the filter parameters and also serve as a quantifying platform to compare the performance of the proposed filter with other non-linear enhancement techniques to be used for diverse biomedical image sensors.