Gaussian-Filtered High-Frequency-Feature Trained Optimized BiLSTM Network for Spoofed-Speech Classification.

Voice-controlled devices are in demand due to their hands-free controls. However, using voice-controlled devices in sensitive scenarios like smartphone applications and financial transactions requires protection against fraudulent attacks referred to as "speech spoofing". The algorithms used in spoof attacks are practically unknown; hence, further analysis and development of spoof-detection models for improving spoof classification are required. A study of the spoofed-speech spectrum suggests that high-frequency features are able to discriminate genuine speech from spoofed speech well. Typically, linear or triangular filter banks are used to obtain high-frequency features. However, a Gaussian filter can extract more global information than a triangular filter. In addition, MFCC features are preferable among other speech features because of their lower covariance. Therefore, in this study, the use of a Gaussian filter is proposed for the extraction of inverted MFCC (iMFCC) features, providing high-frequency features. Complementary features are integrated with iMFCC to strengthen the features that aid in the discrimination of spoof speech. Deep learning has been proven to be efficient in classification applications, but the selection of its hyper-parameters and architecture is crucial and directly affects performance. Therefore, a Bayesian algorithm is used to optimize the BiLSTM network. Thus, in this study, we build a high-frequency-based optimized BiLSTM network to classify the spoofed-speech signal, and we present an extensive investigation using the ASVSpoof 2017 dataset. The optimized BiLSTM model is successfully trained with the least epoch and achieved a 99.58% validation accuracy. The proposed algorithm achieved a 6.58% EER on the evaluation dataset, with a relative improvement of 78% on a baseline spoof-identification system.

A novel approach for yoga pose estimation based on in-depth analysis of human body joint detection accuracy.

Virtual motion and pose from images and video can be estimated by detecting body joints and their interconnection. The human body has diverse and complicated poses in yoga, making its classification challenging. This study estimates yoga poses from the images using a neural network. Five different yoga poses, viz. downdog, tree, plank, warrior2, and goddess in the form of RGB images are used as the target inputs. The BlazePose model was used to localize the body joints of the yoga poses. It detected a maximum of 33 body joints, referred to as keypoints, covering almost all the body parts. Keypoints achieved from the model are considered as predicted joint locations. True keypoints, as the ground truth body joint for individual yoga poses, are identified manually using the open source image annotation tool named Makesense AI. A detailed analysis of the body joint detection accuracy is proposed in the form of percentage of corrected keypoints (PCK) and percentage of detected joints (PDJ) for individual body parts and individual body joints, respectively. An algorithm is designed to measure PCK and PDJ in which the distance between the predicted joint location and true joint location is calculated. The experiment evaluation suggests that the adopted model obtained 93.9% PCK for the goddess pose. The maximum PCK achieved for the goddess pose, i.e., 93.9%, PDJ evaluation was carried out in the staggering mode where maximum PDJ is obtained as 90% to 100% for almost all the body joints.

Mobile and wearable technologies for the analysis of Ten Meter Walk Test: A concise systematic review.

Physical issues started to receive more attention due to the sedentary lifestyle prevalent in modern culture. The Ten Meter Walk Test allows measuring the person's capacity to walk along 10 m and analyzing the advancement of various medical procedures for ailments, including stroke. This systematic review is related to the use of mobile or wearable devices to measure physical parameters while administering the Ten Meter Walk Test for the analysis of the performance of the test. We applied the PRISMA methodology for searching the papers related to the Ten Meter Walk Test. Natural Language Processing (NLP) algorithms were used to automate the screening process. Various papers published in two decades from multiple scientific databases, including IEEE Xplore, Elsevier, Springer, EMBASE, SCOPUS, Multidisciplinary Digital Publishing Institute (MDPI), and PubMed Central were analyzed, focusing on various diseases, devices, features, and methods. The study reveals that chronometer and accelerometer sensors measuring spatiotemporal features are the most pertinent in the Gait characterization of most diseases. Likewise, all studies emphasized the close relation between the quality of the sensor's data obtained and the system's ultimate accuracy. In other words, calibration procedures are needed because of the body part where the sensor is worn and the type of sensor. In addition, using ambient sensors providing kinematic and kinetic features in conjunction with wearable sensors and consistently acquiring walking signals can enhance the system's performance. The most common weaknesses in the analyzed studies are the sample size and the unavailability of continuous monitoring devices for measuring the Ten Meter Walk Test.

Statistical analysis by using soft computing methods for seawater biodegradability using ZnO photocatalyst.

Water quality plays a significant role as a key factor in water resource management. The photocatalytic method is widely used for the removal of recalcitrant pollutants present in seawater. Photocatalysis is a cost-effective technology, sustainable, and environmentally friendly treatment process. In the current approach, a batch reactor was utilized experimentally to study the degradation of contaminants present in seawater by utilizing ZnO as a photocatalyst under natural sunlight. The performance of the process was studied by measuring the percentage removal efficiencies of total organic carbon (TOC), chemical oxygen demand (COD), biological oxygen demand (BOD), and biodegradability with respect to photocatalyst dosage, reaction time and pH of the solution. Biodegradability is defined as the ratio of BOD to COD and this parameter significantly removes pollutants from seawater. The higher the biodegradability, the better the performance of the treatment technology. It also significantly reduces the fouling characteristics of seawater during the desalination process. According to experimental values, the maximum percentage removal efficiencies were found to be TOC = 45.6, COD = 65.4, BOD = 20.01% and biodegradability = 0.038 with respect to the initial values of the seawater sample. The response surface methodology based on Box Behnken design (RSM-BBD) and a predictive model based on the MATLAB adaptive neuro-fuzzy inference system (ANFIS) tool were employed for modeling, optimizing, and evaluating the effects of parameters. According to the RSM-BBD and ANFIS models, the determination coefficients were R2 = 0.959 and R2 = 0.99, respectively, which was very close to 1. The maximum percentage removal efficiencies according to the RSM-BBD design were found to be TOC = 40.3; COD = 61.9; BOD = 18.8% and BOD/COD = 0.0390, whereas for the ANFIS model, the maximum reduction were found to be TOC = 46.5; COD = 65.4; BOD = 20.4% and BOD/COD = 0.040. In process optimization, the ANFIS model was shown better prediction than RSM-BBD in the process's optimization.

Comparative studies of RSM Box-Behnken and ANN-Anfis fuzzy statistical analysis for seawater biodegradability using TiO2 photocatalyst.

In this approach, a batch reactor was employed to study the degradation of pollutants under natural sunlight using TiO2 as a photocatalyst. The effects of photocatalyst dosage, reaction time and pH were investigated by evaluating the percentage removal efficiencies of total organic carbon (TOC), chemical oxygen demand (COD), biological oxygen demand (BOD) and biodegradability (BOD/COD). Design Expert-Response Surface Methodology Box Behnken Design (BBD) and MATLAB Artificial Neural Network - Adaptive Neuro Fuzzy Inference system (ANN-ANFIS) methods were employed to perform the statistical modelling. The experimental values of maximum percentage removal efficiencies were found to be TOC = 82.4, COD = 85.9, BOD = 30.9% and biodegradability was 0.070. According to RSM-BBD and ANFIS analysis, the maximum percentage removal efficiencies were found to be TOC = 90.3, 82.4; COD = 85.4, 85.9; BOD = 28.9, 30.9% and the biodegradability = 0.074, 0.080 respectively at the pH 7.5, reaction time 300 min and photocatalyst dosage of 4 g L-1. The study reveals both models found to be well predicted as compared with experimental values. The values of R2 for RSM-BBD (0.920) and for ANFIS (0.990) models were almost close to 1. The ANFIS model was found to be marginally better than that of RSM-BBD.

A fast region-based active contour for non-rigid object tracking and its shape retrieval.

Conventional tracking approaches track objects using a rectangle bounding box. Gait, gesture and many medical analyses require non-rigid shape extraction. A non-rigid object tracking is more difficult because it needs more accurate object shape and background separation in contrast to rigid bounding boxes. Active contour plays a vital role in the retrieval of image shape. However, the large computation time involved in contour tracing makes its use challenging in video processing. This paper proposes a new formation of the region-based active contour model (ACM) using a mean-shift tracker for video object tracking and its shape retrieval. The removal of re-initialization and fast deformation of the contour is proposed to retrieve the shape of the desired object. A contour model is further modified using a mean-shift tracker to track and retrieve shape simultaneously. The experimental results and their comparative analysis concludes that the proposed contour-based tracking succeed to track and retrieve the shape of the object with 71.86% accuracy. The contour-based mean-shift tracker resolves the scale-orientation selection problem in non-rigid object tracking, and resolves the weakness of the erroneous localization of the object in the frame by the tracker.

Spectral-Spatial Features Integrated Convolution Neural Network for Breast Cancer Classification.

Cancer identification and classification from histopathological images of the breast depends greatly on experts, and computer-aided diagnosis can play an important role in disagreement of experts. This automatic process has increased the accuracy of the classification at a reduced cost. The advancement in Convolution Neural Network (CNN) structure has outperformed the traditional approaches in biomedical imaging applications. One of the limiting factors of CNN is it uses spatial image features only for classification. The spectral features from the transform domain have equivalent importance in the complex image classification algorithm. This paper proposes a new CNN structure to classify the histopathological cancer images based on integrating the spectral features obtained using a multi-resolution wavelet transform with the spatial features of CNN. In addition, batch normalization process is used after every layer in the convolution network to improve the poor convergence problem of CNN and the deep layers of CNN are trained with spectral-spatial features. The proposed structure is tested on malignant histology images of the breast for both binary and multi-class classification of tissue using the BreaKHis Dataset and the Breast Cancer Classification Challenge 2015 Datasest. Experimental results show that the combination of spectral-spatial features improves classification accuracy of the CNN network and requires less training parameters in comparison with the well known models (i.e., VGG16 and ALEXNET). The proposed structure achieves an average accuracy of 97.58% and 97.45% with 7.6 million training parameters on both datasets, respectively.