• Korean Journal of Radiology
• /
• v.22 no.12
• /
• pp.2017-2025
• /
• 2021

Objective: To evaluate the accuracy and clinical efficacy of a hybrid Greulich-Pyle (GP) and modified Tanner-Whitehouse (TW) artificial intelligence (AI) model for bone age assessment. Materials and Methods: A deep learning-based model was trained on an open dataset of multiple ethnicities. A total of 102 hand radiographs (51 male and 51 female; mean age ± standard deviation = 10.95 ± 2.37 years) from a single institution were selected for external validation. Three human experts performed bone age assessments based on the GP atlas to develop a reference standard. Two study radiologists performed bone age assessments with and without AI model assistance in two separate sessions, for which the reading time was recorded. The performance of the AI software was assessed by comparing the mean absolute difference between the AI-calculated bone age and the reference standard. The reading time was compared between reading with and without AI using a paired t test. Furthermore, the reliability between the two study radiologists' bone age assessments was assessed using intraclass correlation coefficients (ICCs), and the results were compared between reading with and without AI. Results: The bone ages assessed by the experts and the AI model were not significantly different (11.39 ± 2.74 years and 11.35 ± 2.76 years, respectively, p = 0.31). The mean absolute difference was 0.39 years (95% confidence interval, 0.33-0.45 years) between the automated AI assessment and the reference standard. The mean reading time of the two study radiologists was reduced from 54.29 to 35.37 seconds with AI model assistance (p < 0.001). The ICC of the two study radiologists slightly increased with AI model assistance (from 0.945 to 0.990). Conclusion: The proposed AI model was accurate for assessing bone age. Furthermore, this model appeared to enhance the clinical efficacy by reducing the reading time and improving the inter-observer reliability.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.3
• /
• pp.151-158
• /
• 2024

Although on-device artificial intelligence (AI) has gained attention to diagnosing machine faults in real time, most previous studies did not consider the model retraining and redeployment processes that must be performed in real-world industrial environments. Our study addresses this challenge by proposing an on-device AI-based real-time machine fault diagnosis system that utilizes continual learning. Our proposed system includes a lightweight convolutional neural network (CNN) model, a continual learning algorithm, and a real-time monitoring service. First, we developed a lightweight 1D CNN model to reduce the cost of model deployment and enable real-time inference on the target edge device with limited computing resources. We then compared the performance of five continual learning algorithms with three public bearing fault datasets and selected the most effective algorithm for our system. Finally, we implemented a real-time monitoring service using an open-source data visualization framework. In the performance comparison results between continual learning algorithms, we found that the replay-based algorithms outperformed the regularization-based algorithms, and the experience replay (ER) algorithm had the best diagnostic accuracy. We further tuned the number and length of data samples used for a memory buffer of the ER algorithm to maximize its performance. We confirmed that the performance of the ER algorithm becomes higher when a longer data length is used. Consequently, the proposed system showed an accuracy of 98.7%, while only 16.5% of the previous data was stored in memory buffer. Our lightweight CNN model was also able to diagnose a fault type of one data sample within 3.76 ms on the Raspberry Pi 4B device.

• Convergence Security Journal
• /
• v.23 no.5
• /
• pp.57-64
• /
• 2023

In this study, we analyzed how to steal the target model without training data. Input data is generated using the generative model, and a similar model is created by defining a loss function so that the predicted values of the target model and the similar model are close to each other. At this time, the target model has a process of learning so that the similar model is similar to it by gradient descent using the logit (logic) value of each class for the input data. The tensorflow machine learning library was used as an experimental environment, and CIFAR10 and SVHN were used as datasets. A similar model was created using the ResNet model as a target model. As a result of the experiment, it was found that the model stealing method generated a similar model with an accuracy of 86.18% for CIFAR10 and 96.02% for SVHN, producing similar predicted values to the target model. In addition, considerations on the model stealing method, military use, and limitations were also analyzed.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.31-40
• /
• 2024

In the realm of domestic online fashion platform industry the manual registration of product information by individual business owners leads to inconvenience and reliability issues, especially when dealing with simultaneous registrations of numerous product groups. Moreover, bias is significantly heightened due to the low quality of product images and an imbalance in data quantity. Therefore, this study proposes a ResNet50 model aimed at minimizing data bias through oversampling techniques and conducting multiple classifications for 13 fashion categories. Transfer learning is employed to optimize resource utilization and reduce prolonged learning times. The results indicate improved discrimination of up to 33.4% for data augmentation in classes with insufficient data compared to the basic convolution neural network (CNN) model. The reliability of all outcomes is underscored by precision and affirmed by the recall curve. This study is suggested to advance the development of the domestic online fashion platform industry to a higher echelon.

• Computers and Concrete
• /
• v.33 no.1
• /
• pp.55-75
• /
• 2024

The present study introduces an optimum performance soft computing model for predicting the compressive strength of high-performance concrete (HPC) by comparing models based on conventional (kernel-based, covariance function-based, and tree-based), advanced machine (least square support vector machine-LSSVM and minimax probability machine regressor-MPMR), and deep (artificial neural network-ANN) learning approaches using a common database for the first time. A compressive strength database, having results of 1030 concrete samples, has been compiled from the literature and preprocessed. For the purpose of training, testing, and validation of soft computing models, 803, 101, and 101 data points have been selected arbitrarily from preprocessed data points, i.e., 1005. Thirteen performance metrics, including three new metrics, i.e., a20-index, index of agreement, and index of scatter, have been implemented for each model. The performance comparison reveals that the SVM (kernel-based), ET (tree-based), MPMR (advanced), and ANN (deep) models have achieved higher performance in predicting the compressive strength of HPC. From the overall analysis of performance, accuracy, Taylor plot, accuracy metric, regression error characteristics curve, Anderson-Darling, Wilcoxon, Uncertainty, and reliability, it has been observed that model CS4 based on the ensemble tree has been recognized as an optimum performance model with higher performance, i.e., a correlation coefficient of 0.9352, root mean square error of 5.76 MPa, and mean absolute error of 4.1069 MPa. The present study also reveals that multicollinearity affects the prediction accuracy of Gaussian process regression, decision tree, multilinear regression, and adaptive boosting regressor models, novel research in compressive strength prediction of HPC. The cosine sensitivity analysis reveals that the prediction of compressive strength of HPC is highly affected by cement content, fine aggregate, coarse aggregate, and water content.

• The Journal of Korea Robotics Society
• /
• v.19 no.2
• /
• pp.176-187
• /
• 2024

Recently, automated solutions using collaborative robots have been emerging in various industries. Their primary functions include Pick & Place, Peg in the Hole, fastening and assembly, welding, and more, which are being utilized and researched in various fields. The application of these robots varies depending on the characteristics of the grippers attached to the end of the collaborative robots. To grasp a variety of objects, a gripper with a high degree of freedom is required. In this paper, we propose a service automation system using a multi-degree-of-freedom gripper, collaborative robots, and vision sensors. Assuming various products are placed at a checkout counter, we use three cameras to recognize the objects, estimate their pose, and create grasping points for grasping. The grasping points are grasped by the multi-degree-of-freedom gripper, and experiments are conducted to recognize barcodes, a key task in service automation. To recognize objects, we used a CNN (Convolutional Neural Network) based algorithm and point cloud to estimate the object's 6D pose. Using the recognized object's 6d pose information, we create grasping points for the multi-degree-of-freedom gripper and perform re-grasping in a direction that facilitates barcode scanning. The experiment was conducted with four selected objects, progressing through identification, 6D pose estimation, and grasping, recording the success and failure of barcode recognition to prove the effectiveness of the proposed system.

• Korean Journal of Agricultural Science
• /
• v.50 no.4
• /
• pp.675-696
• /
• 2023

Terpenoids, also referred to as terpenes, are a large family of naturally occurring chemical compounds present in the essential oils extracted from medicinal plants. In this study, a nondestructive methodology was created by combining ATR-FT-IR (attenuated total reflectance-Fourier transform infrared), and Raman spectroscopy for the terpenoids assessment in medicinal plants essential oils from ten different geographical locations. Partial least squares regression (PLSR) and support vector regression (SVR) were used as machine learning methodologies. However, a deep learning based model called as one-dimensional convolutional neural network (1D CNN) were also developed for models comparison. With a correlation coefficient (R²) of 0.999 and a lowest RMSEP (root mean squared error of prediction) of 0.006% for the prediction datasets, the SVR model created for FT-IR spectral data outperformed both the PLSR and 1 D CNN models. On the other hand, for the classification of essential oils derived from plants collected from various geographical regions, the created SVM (support vector machine) classification model for Raman spectroscopic data obtained an overall classification accuracy of 0.997% which was superior than the FT-IR (0.986%) data. Based on the results we propose that FT-IR spectroscopy, when coupled with the SVR model, has a significant potential for the non-destructive identification of terpenoids in essential oils compared with destructive chemical analysis methods.

• Journal of Internet Computing and Services
• /
• v.25 no.2
• /
• pp.49-55
• /
• 2024

This paper proposes PredFeed Net, a neural network model that mimics the food distribution of fish farming experts. Unlike existing food distribution automation systems, PredFeed Net predicts food distribution by learning the food distribution patterns of experts. This has the advantage of being able to learn using only existing environmental data and food distribution records from food distribution experts, without the need to experiment by changing food distribution variables according to the environment in an actual aquarium. After completing training, PredFeed Net predicts the next food ration based on the current environment or fish condition. Prediction of feed ration is a necessary element for automating feed ration, and feed ration automation contributes to the development of modern fish farming such as smart aquaculture and aquaponics systems.

• Journal of IKEEE
• /
• v.28 no.2
• /
• pp.168-173
• /
• 2024

Speaker recognition refers to a technology that analyzes voice frequencies that are different for each individual and compares them with pre-stored voices to determine the identity of the person. Deep learning-based speaker recognition is being applied to many fields, and pet robots are one of them. However, the hardware performance of pet robots is very limited in terms of the large memory space and calculations of deep learning technology. This is an important problem that pet robots must solve in real-time interaction with users. Lightening deep learning models has become an important way to solve the above problems, and a lot of research is being done recently. In this paper, we describe the results of research on lightweight speaker recognition for pet robots by constructing a voice data set for pet robots, which is a specific command type, and comparing the results of models using residuals. In the conclusion, we present the results of the proposed method and Future research plans are described.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.66 no.4
• /
• pp.27-39
• /
• 2024

Soil texture is determined by the proportions of sand, silt, and clay within the soil, which influence characteristics such as porosity, water retention capacity, electrical conductivity (EC), and pH. Traditional classification of soil texture requires significant sample preparation including oven drying to remove organic matter and moisture, a process that is both time-consuming and costly. This study aims to explore an alternative method by developing an AI model capable of predicting soil texture from images of pre-sorted soil samples using computer vision and deep learning technologies. Soil samples collected from agricultural fields were pre-processed using sieve analysis and the images of each sample were acquired in a controlled studio environment using a smartphone camera. Color distribution ratios based on RGB values of the images were analyzed using the OpenCV library in Python. A convolutional neural network (CNN) model, built on PyTorch, was enhanced using Digital Image Processing (DIP) techniques and then trained across nine distinct conditions to evaluate its robustness and accuracy. The model has achieved an accuracy of over 80% in classifying the images of pre-sorted soil samples, as validated by the components of the confusion matrix and measurements of the F1 score, demonstrating its potential to replace traditional experimental methods for soil texture classification. By utilizing an easily accessible tool, significant time and cost savings can be expected compared to traditional methods.