Journal of Ambient Intelligence and Humanized Computing最新文献

To develop rice varieties with better nutritional qualities, it is important to classify rice seeds accurately. Hyperspectral imaging can be used to extract spectral information from rice seeds, which can then be used to classify them into different varieties. The challenges of precise classification increase when there are many classes and few training samples. In this paper, we present a novel method for high-precision Hyperspectral Image (HSI) classification of 90 different classes of rice seeds using ensemble deep learning. Our method first employs band selection techniques to select the optimal hyperspectral bands for rice seed classification. Then, a deep neural network is trained with the selected hyperspectral and RGB data from rice seed images to obtain different models for different bands. Finally, an ensemble of deep learning models is employed to classify rice seed images and improve classification accuracy. The proposed method achieves an overall precision ranging from 92.73 to 96.17% despite a large number of classes and low data samples for each class and with only 15 selected hyperspectral bands. This precision is significantly higher than the state-of-the-art classical machine learning methods like random forest, confirming the effectiveness of the proposed method in classifying hyperspectral images of rice seeds.

As software gets more complicated, diverse, and crucial to people’s daily lives, exploitable software vulnerabilities constitute a major security risk to the computer system. These vulnerabilities allow unauthorized access, which can cause losses in banking, energy, the military, healthcare, and other key infrastructure systems. Most vulnerability scoring methods employ Natural Language Processing to generate models from descriptions. These models ignore Impact scores, Exploitability scores, Attack Complexity and other statistical features when scoring vulnerabilities. A feature vector for machine learning models is created from a description, impact score, exploitability score, attack complexity score, etc. We score vulnerabilities more precisely than we categorize them. The Decision Tree Regressor, Random Forest Regressor, AdaBoost Regressor, K-nearest Neighbors Regressor, and Support Vector Regressor have been evaluated using the metrics explained variance, r-squared, mean absolute error, mean squared error, and root mean squared error. The tenfold cross-validation method verifies regressor test results. The research uses 193,463 Common Vulnerabilities and Exposures from the National Vulnerability Database. The Random Forest regressor performed well on four of the five criteria, and the tenfold cross-validation test performed even better (0.9968 vs. 0.9958).

In this paper, we have proved fixed point results for a pair of soft fuzzy maps in complete ordered soft metric spaces. We have also given some useful corollaries to our main result along with examples. Moreover, the application is also presented in this communication to show the validity of new results.

Abstract

Video summarization is an emerging research field. In particular, static video summarization plays a major role in abstraction and indexing of video repositories. It extracts the vital events in a video such that it covers the entire content of the video. Frames having those important events are called keyframes which are eventually used in video indexing. It also helps in giving an abstract view of the video content such that the internet users are aware of the events present in the video before watching it completely. The proposed research work is focused on efficient static video summarization by extracting various visual features namely color, texture and shape features. These features are aggregated and clustered using a Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. In order to produce good video summary by clustering, the parameters of DBSCAN algorithm are optimized by using a meta heuristic population based optimization called Artificial Algae Algorithm (AAA). The experimental results on two public datasets namely VSUMM and OVP dataset show that the proposed Static Video Summarization with Multi-objective Constrained Optimization (SVS_MCO) achieves better results when compared to existing methods.

The rapid development of multimedia information processing technology provides development opportunities for digitization in sports, among which motion capture technology, as the latest achievement of multimedia information processing technology, has gradually gained the attention of scholars and started to be used for visualization of sports movements. Therefore, this paper introduces a monocular video motion capture method and optimizes it for the problems of reconstructing human movements such as floating, ground penetration and sliding, which provides a technical path for the specific application of motion capture technology in the field of sports training and also provides a technical guarantee for the visualization of sports training movements. Introduced a new motion capture optimization method. This method captures human motion trajectories from monocular videos, and trajectory operations combine human pose estimation and physical constraints. The proposed method uses foot contact judgment to obtain foot contact events for each motion frame. Then, it optimizes the overall body motion trajectory of the key points based on the obtained contact conditions, making the generated motion visually closer to reality. This article proposes LiteHumanPose Net with a inference speed of up to 22FPS, and conducts experimental analysis and comparison of several popular pose estimation methods from the perspectives of frame rate and average accuracy, such as Sim pleBaseline, HRNet, and Hourglass Net. LiteHumanPose Net outperforms Hourglass Net in terms of frame rate and accuracy, while HRNet has high accuracy due to its multiple parameters but low frame rate. The LiteHumanPose network proposed in this article has a good balance between accuracy and frame rate, and has obvious landing advantages.

Coronavirus belongs to the family of Coronaviridae. It is responsible for COVID-19 communicable disease, which has affected 213 countries and territories worldwide. Researchers in computational fields have been active in proposing techniques to filter the information and recommendations about this disease and provide surveillance in controlling this outbreak. Researchers used Chest X-ray images, abdominal Computed Tomography scans, and Tweet datasets for building machine learning and deep learning-based models for COVID-19 predictions and forecasting purposes. Accuracy, sensitivity, specificity, precision, and F1-measure are the five primary evaluation criteria researchers employ to evaluate the quality of their study. This article summarises research works on COVID-19 based on machine learning and deep learning models. The analysis of these research works, along with their limitations and source of datasets, will give a quick start for future research to arrive at a defined direction.

Crowds can lead up to severe disasterous consequences resulting in fatalities. Videos obtained through public cameras or captured by drones flying overhead can be processed with artificial intelligence-based crowd analysis systems. Being a hot area of research over the past few years, the goal is not only to identify the presence of crowds but also to predict the probability of crowd-formation in order to issue timely warnings and preventive measures. Such systems will significantly reduce the probablity of the potential disasters. Developing effective systems is a challenging task, especially due to factors such as naturally occuring diverse conditions, variations in people or background pixel areas, noise, behaviors of individuals, relative amounts/distributions/directions of crowd movements, and crowd building reasons. This paper proposes an infrared video processing system based on U-Net convolutional neural network for crowd monitoring in infrared video frames to help estimate the people crowd with normal or abnormal trends. The proposed U-Net architecture aims to efficiently extract crowd features, achieve sufficient people marking-up accuracy, competitively with optimal network configurations in terms of the depth and number of filters to consequently minimise the number of coefficients. For further faster processing, hardware resources/implementation area savings, and lower power, the optimized network coefficients measured are represented in Canonic-Signed Digit with minimal number of nonzero (± 1) digits, minimizing the number of underlying shift-add/subtract operations of all multipliers. The achieved significantly reduced computational cost makes the proposed U-Net effectively suitable for resource-constrained and low power applications.

Sign language recognition is an important social issue to be addressed which can benefit the deaf and hard of hearing community by providing easier and faster communication. Some previous studies on sign language recognition have used complex input modalities and feature extraction methods, limiting their practical applicability. This research aims to compare two custom-made convolutional neural network (CNN) models for recognizing American Sign Language (ASL) letters from A to Z, and determine which model performs better. The proposed models utilize a combination of CNN and Softmax activation function, which are powerful and widely used classification methods in the field of computer vision. The purpose of the proposed study is to compare the performance of two specially created CNN models for identifying 26 distinct hand signals that represent the 26 English alphabets. The study found that Model_2 had better overall performance than Model_1, with an accuracy of 98.44% and F1 score 98.41%. However, the performance of each model varied depending on the specific label, suggesting that the choice of model may depend on the specific use case and the labels of interest. This research contributes to the growing field of sign language recognition using deep learning techniques and highlights the importance of designing custom models.

There have been many connections between medical science and artificial intelligence in recent years. Many problems arise with the integrity of communication. Cardiac arrhythmia, carried out using artificial intelligence methods, is one of the most dangerous diseases in the field of prevention. Topics introduced in artificial intelligence are the automatic selection of balancing and classification algorithms. In this study, metrics for machine learning algorithm selection are presented. The first problem is the problem of choosing the best balancing algorithm to balance the data sets, introduced as triangle rate (TR). The second issue to be studied is selecting the best automatic classification algorithm. The third action was to use a scoring algorithm to predict sinus and non-sinus arrhythmias. The heptagonal reinforcement learning (HRL) achieved results competitive with standard algorithms by combining three types of algorithms. The data used in this study was a 12-lead electrocardiogram (ECG) database of arrhythmias. The number of patients examined in this dataset is 10,646. The HRL algorithm has improved the previous algorithms by 5%, achieving 86% cardiac arrhythmia prediction.