Accurate bacterial identification is essential in various research fields. Traditional methods, culturing bacteria and phenotypic analysis are time-consuming and laborious while the molecular technique using 16S rRNA gene sequence offers a faster and more precise approach. This study investigated the specificity of universal 16S rRNA primers for bacterial identification. Two-step in silico analysis was conducted using Vibrio parahaemolyticus (GenBank accession EU636231) as a model. First, ten universal primer sets were evaluated for their in silico amplification against the target sequence using BioEdit software to assess their ability to specifically bind to the desired region of the 16S rRNA gene. Second, the in silico amplified products from each primer set were then subjected to a BlastN search against the NCBI database to determine the specificity of the amplified sequence. All primers exhibited some degree of specificity during the in silico amplification step. However, significant variations were observed in BlastN results. Notably, primers targeting longer 16S rRNA regions (>500 bp) demonstrated greater concert in achieving species-level resolution. Four primer sets (20F-1500R, 27F-1492R, V1–V3, 8F-534R) perfectly matched the target and showed high identity (>99.5
It is important to always monitor the health of cattle, especially calves, and the frequency of observation increases with environmental changes in addition to once a day. In addition, calves tend to be more susceptible to infectious diseases because of their immature immune systems. Therefore, rearing management is extremely important. And the number of dairy cattle-keeping households and the total number of cattle are decreasing, while the number of cattle per household is increasing, indicating that management is becoming larger in scale. In this study, we proposed the development of a health management system by analyzing calf behavior using a 3D camera. Experiments were conducted at the Sumiyoshi Field of Miyazaki University to confirm the effectiveness of the proposed method.
This paper describes a deep learning method such as the long short-term memory (LSTM), bidirectional LSTM (Bi-LSTM), gated recurrent unit (GRU), bidirectional gated recurrent unit (Bi-GRU) to design for forecasting in terms of solving this problem. The data for the study was obtained on the Eurostat website from January 2017 to June 2022, and the data for monthly gas consumption used the technique of walk-forward validation (WFV) to evaluate the generalization ability prediction. Based on error scores, the study results compared the accuracy rates of the four models’ approaches show that: the Bi-GRU model was the best prediction for Bulgaria, Hungary, and Romania, the LSTM model best approached Slovakia, and the GRU model was the best suitable for Poland and Czech. The predicted methods for effectively managing the natural gas resource are crucial for enhancing gas consumption’s efficiency and reducing its impact on the environment. In addition, the result may contribute to stakeholders taking the right decision for energy planning.
Given the extensive rice consumption across numerous nations, rice stands as one of the most cultivated crops worldwide. Understanding the stages of rice cultivation within the proposed system is essential for successful rice growth and for gaining knowledge about various diseases affecting rice. Researchers have recently devised several diagnostic techniques for identifying rice leaf diseases. A convolutional neural network (CNN) based classification approach has been formulated to categorize diseases using images of rice leaves collected from rice fields. Identifying rice leaf diseases is difficult due to their diverse nature, each exhibiting unique traits. In this paper, a system for classification of Rice Leaf Diseases is proposed. Two pre-trained models, VGG16 and MobileNet, are used as classifiers. In this system, two datasets are used for the classifier. In Datasets, rice leaf diseases are classified as Bacterial Leaf Blight, Brown Spot, Leaf Smut, Leaf Blast, Sheath Blight, and Healthy. As an evaluation, the performance of the two models is compared.
The technology of cloud computing is becoming more dynamic every day. Providing IT services to academic institutions and other commercial businesses has many advantages. The cloud-based Learning Management System (LMS) has been increasingly used in Comprehensive Universities in developing countries. Lecturers are able to post lesson plans, homework assignments, and exams on the cloud server, which gives students access to all the course materials from home or at university. As noticeable, cloud providers face a variety of security challenges such as risks, threats and vulnerabilities. A cloud provider in an academic institution needs to understand and reduce cloud security risks, improve security and increase confidence in cloud services. Thus, a risk management model is essential for securing cloud-based LMS in academic institutions. This paper proposes a risk management model based on fuzzy rule-based logic for mitigation of security concerns in risks for cloud-based LMS in academic institutions. It identifies and characterizes the critical risk factors associated with insider threats prevalent in a cloud-based LMS environment. It presents the process of collecting and preprocessing log data from server logs and Moodle application logs to extract relevant features indicative of these security risks. It creates a model based on fuzzy rules to evaluate and measure the risk levels. The output of this model is a risk level that provides a nuanced assessment of the security posture of the cloud-based LMS.This proposed model seeks to improve educational institutions’ cloud cybersecurity posture by fortifying LMS platforms’ security resistance against insider threats.
A well-maintained library is a valuable resource for educational institutions and research facilities. Efficient management of libraries requires a reliable system for cataloging books. This paper evaluates four prominent edge detectors - Canny, Sobel, Prewitt and Roberts, analyzing their performance using metrics such as mean square error (MSE), root mean square error (RMSE) and peak signal-to-noise ratio (PSNR). Why we study on the evaluation of edge detectors initiated from the studying of the University of Information Technology’s Library, which houses over 100,000 books. We aim to address the challenges of efficiently cataloging and retrieving books in the digital library of this extensive collection to collaborate with an IT research center for future improvements. In these days, optimizing edge detection for book spine recognition is essential for efficient cataloging, inventory management and quick retrieval in digital library applications. Thus, this paper presents a comparative analysis of the edge detection techniques to support in solving the challenges of a digital library in future. These techniques are within computer vision and image processing to enhance book spine detection accuracy, reduce manual labor and minimize errors in robotic libraries. Our evaluation results indicate that the Canny edge detector consistently outperforms the others, demonstrating superior performance with lower MSE, RMSE and higher PSNR values. Integrating the Canny edge detector into existing systems will enhance book organization and retrieval in the digital library. This accurate and noise-resistant detector addresses book spine segmentation issues by solving cataloging inefficiencies and improving overall management.
This paper adopts YOLOv8, a cutting-edge deep learning networks, to segment road scenes, identify and enhance roadside, stairs, and tactile paving, and display improved road signs and Tactile paving through AR glasses. Key to the visual guidance system is the accurate identification and positioning of road signs, allowing for the enlargement or color highlighting of these signs, and providing auditory prompts to the user. This multi-modal approach aims to mitigate the effects of visual impairments, thereby enhancing the travel experience and safety for individuals with low vision.
In this paper, we propose a novel method for analyzing piglet calls that includes an efficient noise reduction technique tailored for noisy pig farming environments. Our approach adopts analysis of the inaudible frequency ranges, where environmental noise levels are low, combined with Non-negative Matrix Factorization (NMF) based noise reduction. We constructed a Random Forest classifier using sixteen acoustic features as an acoustic event detector. The experiment revealed that the three acoustic features, which are F0, ∆MFCC, and SpBandwidth, were particularly important, as they were able to distinguish between the squeals, litter calls, and environmental noise of the three piglet species 98.9
The performance evaluation of autonomous navigation systems plays an important role in engineering applications, but there are few researches on the general capability evaluation of autonomous navigation systems. To solve this problem, this paper proposes a general capability evaluation method of autonomous navigation system based on similarity calculation, which aims to effectively measure the performance of autonomous navigation system in complex environment. This method provides a scientific decision basis for the design optimization and decision support of autonomous navigation systems, and plays an important role in improving the general capability of autonomous navigation systems and coping with diversified task challenges.
Human beings are prone to limb fractures or physical sports injuries during exercise, and patients with sports injuries will no longer be followed up after treatment to relieve pain, resulting in sequelae of injuries caused by patients, and when through a complete rehabilitation assessment, not only can improve the sequelae, but also strengthen the body. However, it is very important to effectively record the patient’s rehabilitation process and provide medical staff with a complete rehabilitation record so that the physician can grasp the patient’s training results. In this paper, the node sensor is used to detect gait, which is divided into walking, running and rope skipping, and the characteristics are expressed by the pedometer calculation of the number of steps and frequency domain detection, and the similarity is obtained by using the dynamic time warp time (DTW) calculus to determine the patient’s gait abnormality through real-time comparison between the similarity and the normal gait, and the similarity of the training can reach 8
Rumination is a critical indicator of a cow’s physiological state, making it a valuable metric for managing cow health and predicting calving. Traditional human observation of rumination behaviour is time consuming and impractical for continuous monitoring, and sensor-based identification can be stressful for the cows due to the need for attachment. To address these challenges, this study introduces a non-contact method for identifying cow rumination. The proposed approach involves capturing video footage of multiple cows from above, recognizing cow regions, and validating the method’s effectiveness through experiments. Specifically, we utilize optical flow and frame-to-frame subtraction methods to extract moving cow regions from the recorded videos. From this data, we derive 11 features and employ a Support Vector Machine (SVM) for classification. Training the SVM with label features resulted in a test data identification accuracy of approximately 60
Physical therapists who instruct rehabilitation for their patients and staffs in welfare facilities who help residents take bath and do other things both face a common troublesome problem on managing the pain that the care receivers claim or causes violation without any notification. Because many methods depend on the self-report by care receivers that is not reliable in most cases (they tend to exaggerate the anxiety for pain). Past research on pain, on the other hand, have clarified facial expressions were reliable enough than words but at the same time they had large individual difference in the location of pain facial expression. This paper reports a method of predicting where pain facial expression appears only by taking an image of face with neutral expression and analyzing it with respect to permanent wrinkles. Successful prediction of pain facial expression enables for care givers to prevent pain-inducing care from continuing unintendedly.
This study explores the potential of the MVP (Musical Pitch Visualization Perception) support system integrated with Google Glass, designed to enhance wind instrumentalists’ performance. The system aims to improve intonation accuracy, reduce cognitive load, and provide flexibility in visual and physical orientation during performance. While previous research has primarily relied on subjective evaluations and qualitative methods, this study employs OpenPose for motion capture to obtain objective data on performers’ body movements. An experiment was conducted where a saxophonist played a B-flat major scale while keeping pace with a metronome, using both a conventional tuner and the MVP system. The results indicated that the MVP system enabled more fluid and detailed body movements, suggesting an increase in the performer’s freedom of movement and more precise acquisition of metronome information. Despite the promising findings, the study acknowledges limitations due to the small sample size and recommends further research with more participants and dedicated motion capture equipment for more robust analysis.
In this paper, we propose a method for simultaneously estimating each TDOA in a multi-loudspeaker environment using GCC-PHAT. Conventional TDOA estimation systems typically generate sound exclusively from a single loudspeaker, even in the presence of multiple loudspeakers, and estimate the TDOA individually. However, the accuracy degrades significantly when sound sources are present simultaneously from multiple loudspeakers. In this paper, we propose an arrival time estimation method based on the GCC-PHAT method to solve this problem. The proposed method allows detection with mixed sounds, which is not possible with conventional methods. The results of evaluation experiments using computer simulations confirm that the proposed method can estimate the arrival time of each loudspeaker in an environment where sound is played from two loudspeakers. In addition, as a result of changing parameters such as embedding intensity, ratio of frequencies, and time, it was confirmed that under certain conditions, the arrival time estimation and distance estimation errors were smaller than those of the conventional method. As a result, this proposed method reduces the arrival time estimation error and distance estimation error to zero for the two types of sources.
Metaheuristic algorithms have shown strong advantages in solving real application problems, and excellent metaheuristic algorithms have emerged in recent years. The Willow Catkin Optimization (WCO) algorithm is a novel metaheuristic algorithm for modeling willow reproductive behavior, which has a simple structure, is easy to implement, and has excellent search capability. In this paper, the WCO algorithm is employed to address the coverage problem of wireless sensor networks (WSNs) in 2D and 3D environments and the experimental results are compared with other classical algorithms and lates proposed algorithms, and the experimental results show that the WCO algorithm has excellent performance in solving the node coverage problem.
Transport and logistics operated by unmanned aerial vehicles, commonly known as drones, have attracted attention in recent years for their potential to revolutionize the transport industries. For example, Amazon was the first to use drones to deliver goods. Several distribution companies have since been working on similar services. To find effective delivery routes by using vehicles and drones, a flying sidekick traveling salesman problem has firstly been formulated. This problem constructs a delivery route by using a single drone and a single vehicle. In further formulation of this problem, vehicle routing problem with drones (VRPD), in which several numbers of drones and vehicles deliver goods to customers, has been formulated. We have already proposed a solution search method based on chaotic neurodynamics for VRPD. In this article, to further achieve effective search capability for VRPD, we change neural codings of chaotic neural networks according to the combinations of neighborhood operations. Experimental results show that our proposed method exhibits better objective function values than those by the conventional chaotic search method.
This study proposes an automatic labeling method for liver regions in Computed Tomography (CT) images, addressing the time-consuming nature of manual medical image labeling. The approach utilizes edge detection to identify overall edge distributions in CT images, constrains the region of interest to locate characteristic liver areas, and performs object segmentation to generate liver masks for labeling. The method tackles key challenges in liver CT image analysis, including image noise, blurred edge features, and minimal pixel differences in suspected tumor areas. It employs Gaussian blur for noise reduction, followed by brightness and contrast adjustments to enhance edge characteristics. The watershed algorithm is then applied to segment the complete liver contour. Performance evaluation using approximately 17,000 images from the LiTS datasets demonstrated an average accuracy of 75
This paper proposes a mask segmentation and evaluation method based on Mask Scoring R-CNN, which combines Mask R-CNN and the mask evaluation mechanism. We adopt Mask R-CNN to extract accurate target masks from images, and propose a mask evaluation mechanism to evaluate the quality of the generated masks. By presenting mask scoring, we are able to accurately measure the accuracy and completeness of masks, thereby improving the quality of mask segmentation. In order to ensure the accuracy and completeness of segmentation, we have optimized Mask Scoring R-CNN by adding a loss function that can be used for multi-category image segmentation. This loss function can improve the segmentation task . The evaluation index IoU is optimized and designed to improve the original mask quality score while maintaining a high accuracy of the segmented images.
This paper presents a method for analyzing Japanese Diet deliberations video footage to determine if a speaker reads from a script while speaking. The study focuses on classifying script reading faces as a preprocessing step for future research. Four main classification methods are proposed: (1) using the success or failure of face detection, (2) utilizing head pose angle based on the PnP problem, (3) employing head pose angle based on a regression model, and (4) using an image recognition model. Twelve videos from a Video Retrieval System for Diet Deliberations were used, and 30-s speaking scenes were extracted to evaluate the proposed approaches. The results indicate that approach (4) achieved the highest classification accuracy.
Although mechanical weighing machines specialized for pigs are normally used to measure animal weights on pig farms, guiding those animals onto the weighing machines is a difficult and unpleasant chore that normally takes 20 seconds to complete after the pig is positioned on the load cell. Additionally, pig weighing machines are prone to mechanical breakdowns and are unpleasant to repair because of the farm residue they collect. Therefore, the development of a more practical and robust weight measurement apparatus is desired. In the present paper, we report on a handheld weight estimation device using a red-green-blue-depth (RGB-D) sensor with a laser slit system. An RGB-D sensor system is used to estimate pig weights, and the laser slit is used to align the measurement direction of the RGB-D sensor along the pig’s body. In operation, this system captures a depth image of the pig under examination and then uses it to create three-dimensional (3D) data. The 3D body parameters are extracted using image processing, and those extracted 3D body parameters for the pig are then fed into a random forest algorithm to produce the weight estimation. We also report on experimental results that demonstrate the reliability of our pig weight estimation system and its suitability for practical use.
