Parameter-free fuzzy histogram equalisation with illumination preserving characteristics dedicated for contrast enhancement of magnetic resonance images

Gudermannian neural networks using the optimization procedures of genetic algorithm and active set approach for the three-species food chain nonlinear model

The present study is to investigate the Gudermannian neural networks (GNNs) using the optimization procedures of genetic algorithm and active-set approach (GA-ASA) to solve the three-species food chain nonlinear model. The three-species food chain nonlinear model is dependent upon the prey populations, top-predator, and specialist predator. The design of an error-based fitness function is presented using the sense of the three-species food chain nonlinear model and its initial conditions. The numerical results of the model have been obtained by exploiting the GNN-GA-ASA. The obtained results through the GNN-GA-ASA have been compared with the Runge-Kutta method to substantiate the correctness of the designed approach. The reliability, efficacy and authenticity of the proposed GNN-GA-ASA are examined through different statistical measures based on single and multiple neurons for solving the three-species food chain nonlinear model.

An Automated In-Depth Feature Learning Algorithm for Breast Abnormality Prognosis and Robust Characterization from Mammography Images Using Deep Transfer Learning

BACKGROUND

Diagnosing breast cancer masses and calcification clusters have paramount significance in mammography, which aids in mitigating the disease's complexities and curing it at early stages. However, a wrong mammogram interpretation may lead to an unnecessary biopsy of the false-positive findings, which reduces the patient's survival chances. Consequently, approaches that learn to discern breast masses can reduce the number of misconceptions and incorrect diagnoses. Conventionally used classification models focus on feature extraction techniques specific to a particular problem based on domain information. Deep learning strategies are becoming promising alternatives to solve the many challenges of feature-based approaches.

METHODS

This study introduces a convolutional neural network (ConvNet)-based deep learning method to extract features at varying densities and discern mammography's normal and suspected regions. Two different experiments were carried out to make an accurate diagnosis and classification. The first experiment consisted of five end-to-end pre-trained and fine-tuned deep convolution neural networks (DCNN). The in-depth features extracted from the ConvNet are also used to train the support vector machine algorithm to achieve excellent performance in the second experiment. Additionally, DCNN is the most frequently used image interpretation and classification method, including VGGNet, GoogLeNet, MobileNet, ResNet, and DenseNet. Moreover, this study pertains to data cleaning, preprocessing, and data augmentation, and improving mass recognition accuracy. The efficacy of all models is evaluated by training and testing three mammography datasets and has exhibited remarkable results.

RESULTS

Our deep learning ConvNet+SVM model obtained a discriminative training accuracy of 97.7% and validating accuracy of 97.8%, contrary to this, VGGNet16 method yielded 90.2%, 93.5% for VGGNet19, 63.4% for GoogLeNet, 82.9% for MobileNetV2, 75.1% for ResNet50, and 72.9% for DenseNet121.

CONCLUSIONS

The proposed model's improvement and validation are appropriated in conventional pathological practices that conceivably reduce the pathologist's strain in predicting clinical outcomes by analyzing patients' mammography images.

Detection of Defects in Geomembranes Using Quasi-Active Infrared Thermography

High-density polyethylene geomembranes are employed as covers for the sewage treatment lagoons at Melbourne Water Corporation's Western Treatment Plant, to harvest the biogas produced during anaerobic degradation, which is then used to generate electricity. Due to its size, inspecting the cover for defects, particularly subsurface defects, can be challenging, as well as the potential for the underside of the membrane to come into contact with different substrates, viz. liquid sewage, scum (consolidated solid matter), and biogas. This paper presents the application of a novel quasi-active thermography inspection method for subsurface defect detection in the geomembrane. The proposed approach utilises ambient sunlight as the input thermal energy and cloud shading as the trigger for thermal transients. Outdoor laboratory-scale experiments were conducted to study the proposed inspection technique. A pyranometer was used to measure the intensity of solar radiation, and an infrared thermal camera was used to measure the surface temperature of the geomembrane. The measured temperature profile was analysed using three different algorithms for thermal transient analysis, based on (i) the cooling constant from Newton's law of cooling, (ii) the peak value of the logarithmic second derivative, and (iii) a frame subtraction method. The outcomes from each algorithm were examined and compared. The results show that, while each algorithm has some limitations, when used in combination the three algorithms could be used to distinguish between different substrates and to determine the presence of subsurface defects.

Image Watermarking Approach Using a Hybrid Domain Based on Performance Parameter Analysis

In today’s scenario, image watermarking has been an integral part in various multimedia applications. Watermarking is the approach for adding additional information to the existing image to protect the data from modification and to provide data integrity. Frequency transform domain techniques are complex and costly and degrade the quality of the image due to less embedding of bits. The proposed work utilize the original DCT method with some modifications and applies this method on frequency bands of DWT. Furthermore, the output is used in combination with a pixel modification method for embedding and extraction. The proposed outcome is the improvement of performance achieved in terms of time, imperceptibility, and robustness.