Taiwan Association of Engineering and Technology Innovation: E-Journals
Not a member yet
887 research outputs found
Sort by
Estimating Classification Accuracy for Unlabeled Datasets Based on Block Scaling
Get PDFThis paper proposes an approach called block scaling quality (BSQ) for estimating the prediction accuracy of a deep network model. The basic operation perturbs the input spectrogram by multiplying all values within a block by , where is equal to 0 in the experiments. The ratio of perturbed spectrograms that have different prediction labels than the original spectrogram to the total number of perturbed spectrograms indicates how much of the spectrogram is crucial for the prediction. Thus, this ratio is inversely correlated with the accuracy of the dataset. The BSQ approach demonstrates satisfactory estimation accuracy in experiments when compared with various other approaches. When using only the Jamendo and FMA datasets, the estimation accuracy experiences an average error of 4.9% and 1.8%, respectively. Moreover, the BSQ approach holds advantages over some of the comparison counterparts. Overall, it presents a promising approach for estimating the accuracy of a deep network model
Simulation Study on a New Hybrid Autonomous Underwater Vehicle with Elevators
Get PDFThis study aims to design a new hybrid twin autonomous underwater vehicle (HTAUV) consisting of dual cylinder hulls and analyze its pitching motion. The kinematic model for the HTAUV is established, followed by the execution of hydrodynamic simulation CFD of the HTAUV using Ansys Fluent. These simulations are conducted to obtain the hydrodynamic force equation of the HTAUV, which relates to the deflection angle of the elevator. Through the motion simulation of the HTAUV, under the same net buoyancy condition, notable differences emerge when the elevator is deflected. Specifically, parameters such as gliding speed, gliding angle, and pitch angle of the HTAUV are larger when the elevator is deflected, as compared to cases where no deflection is applied
On the Estimation of the Mission Performance Index of Unmanned Surface Vehicles Based on the Mission Coverage Area
Get PDFFor mission planning and replanning of multiple unmanned surface vehicles (USVs), it is important to estimate each USV’s mission performance in terms of sea surveillance (e.g., illegal ship control). In this study, a mission performance index (MPI) is proposed based on the mission coverage area for estimating the USVs’ mission performance of illegal ship control. The penalty value is considered in the MPI calculation procedure owing to the track-off of the USV. In addition, the USV simulation is conducted under illegal ship control, and the MPI is calculated based on changing the mission coverage area. The results show that the MPI increases with the path width of the mission coverage area
A Robust Technique for Detection, Diagnosis, and Localization of Switching Faults in Electric Drives Using Discrete Wavelet Transform
Get PDFDetection, diagnosis, and localization of switching faults in electric drives are extremely important for operating a large number of induction motors in parallel. This study aims to present the design and development of switching fault detection, diagnosis, and localization strategy for the induction motor drive system (IMDS) by using a novel diagnostic variable that is derived from discrete wavelet transform (DWT) coefficients. The distinctiveness of the proposed algorithm is that it can identify single/multiple switch open and short faults and locate the defective switches using a single mathematical computation. The proposed algorithm is tested by simulation in MATLAB/Simulink and experimentally validated using the LabVIEW hardware-in-the-loop platform. The results demonstrate the robustness and effectiveness of the proposed technique in identifying and locating faults
An Effective Supervised Machine Learning Approach for Indian Native Chicken’s Gender and Breed Classification
Get PDFThis study proposes a computer vision and machine learning (ML)-based approach to classify gender and breed in native chicken production industries with minimal human intervention. The supervised ML and feature extraction algorithms are utilized to classify eleven Indian chicken breeds, with 17,600 training samples and 4,400 testing samples (80:20 ratio). The gray-level co-occurrence matrix (GLCM) algorithm is applied for feature extraction, and the principle component analysis (PCA) algorithm is used for feature selection. Among the tested 27 classifiers, the FG-SVM, F-KNN, and W-KNN classifiers obtain more than 90% accuracy, with individual accuracies of 90.1%, 99.1%, and 99.1%. The BT classifier performs well in gender and breed classification work, achieving accuracy, precision, sensitivity, and F-scores of 99.3%, 90.2%, 99.4%, and 99.5%, respectively, and a mean absolute error of 0.7
Shoe Last Customization: A Systematic Review
Get PDFIn recent years, there is an increase in research into shoe last customization and topic analysis methods. The work aims to systematically review the literature on the customization of shoe lasts. The method used in this work is to perform a five-phase systematic review algorithm. Data on the research performed are extracted and synthesized from each study: main research objectives, authors, date of publication, journal, or conference in which the article was published, and the quality of each article. The studies included in the review are published between 2018 and 2022. The results of the review are nineteen papers about the process of customization of the shoe last. The conclusions of the analysis indicate that the quality of research has not changed over time, in 2020 there was a decrease in work. Most often, researchers analyze the impact of anthropometric factors on the correct shoe last modeling and methods of shoe last parameterization
A Hidden Semi-Markov Model for Predicting Production Cycle Time Using Bluetooth Low Energy Data
Get PDFThis study proposes a statistical model to characterize the temporal characteristics of an entire production process. The model utilizes received signal strength indicator (RSSI) data obtained from a Bluetooth low energy (BLE) network. A hidden semi-Markov model (HSMM) is formulated based on the characteristics of the production process, and the forward-backward algorithm is employed to re-estimate the probability distribution of state durations. The proposed method is validated through numerical, simulation, and real-world experiments, yielding promising results. The results show that the Kullback-Leibler divergence (KLD) score of 0.1843, while the simulation achieves an average vector distance score of 0.9740. The real-time experiment also shows a reasonable accuracy, with an average HSMM estimated throughput time of 30.48 epochs, compared to the average real throughput time of 33.99 epochs. Overall, the model serves as a valuable tool for predicting the cycle time and throughput time of a production line
Mapping and Change Assessment of Captive Limestone Mining Areas Using Landsat-5/8 Images
Get PDFLimestone is a non-metallic mineral extensively used in cement manufacturing and construction sector. Extensive mineral mining processes impact the environment. The study aims to map and evaluate the limestone mining area change at the Yerraguntla industrial zone in the YSR district of Andhra Pradesh, India. The normalized difference vegetation index (NDVI) and modified soil-adjusted vegetation index (MSAVI) are computed from the Landsat-5/8 images using Quantum GIS (QGIS) software. Experimental results show that the limestone mining area increases from 307 ha to 469.92 ha during 2005-2019. NDVI method is more effective than MSAVI in change assessment of limestone mining areas with overall accuracy of 87.75 % and 79.49 % and kappa coefficient of 0.89 and 0.62 respectively in 2019. The finding is compared with industry field survey reports (487.10 ha). This study contributes to the limestone mining industry management in developing a land-environmental management plan for the long-term sustainability of limestone mining
Effectiveness of Silica Fume Eggshell Ash and Lime Use on the Properties of Kaolinitic Clay
Get PDFThe study aims to investigate the properties of kaolinitic clay using silica fume, eggshell ash, and lime. The experiment employs varying amounts of silica fume (2%, 4%, and 6%), eggshell ash, lime (3%, 6%, and 9%), and combinations of silica fume, eggshell ash, and lime, which are cured for 1, 7, 14, and 30 days. The investigated properties of the soils include the improvement of Atterberg limits, maximum dry density (MDD), optimum moisture content (OMC), specific gravity, compressive strength, morphology characteristics, and chemical compositions. The results reveal that the optimal application of these materials can be achieved at 6% silica fume, 6% eggshell ash, and 9% lime mixture into the soils and increase the shear strength by as much as 88.74% at 30 days of curing
Non-Facial Video Spatiotemporal Forensic Analysis Using Deep Learning Techniques
Get PDFDigital content manipulation software is working as a boon for people to edit recorded video or audio content. To prevent the unethical use of such readily available altering tools, digital multimedia forensics is becoming increasingly important. Hence, this study aims to identify whether the video and audio of the given digital content are fake or real. For temporal video forgery detection, the convolutional 3D layers are used to build a model which can identify temporal forgeries with an average accuracy of 85% on the validation dataset. Also, the identification of audio forgery, using a ResNet-34 pre-trained model and the transfer learning approach, has been achieved. The proposed model achieves an accuracy of 99% with 0.3% validation loss on the validation part of the logical access dataset, which is better than earlier models in the range of 90-95% accuracy on the validation set