• Journal of The Korean Association of Information Education
• /
• v.1 no.1
• /
• pp.28-37
• /
• 1997

This paper presents a new neural network based on fuzzy set and its application to invariant character recognition. The fuzzy neural network consists of five layers. The results of simulation show that the network can recognize characters in the case of distortion, translation, rotation and different sizes of handwritten characters and even with noise(8${\sim}$30%)). Translation, distortion, different sizes and noise are achieved by layer L2 and rotation invariant by layer L5. The network can recognize 108 examples of training with 100% recognition rate when they are shifted in eight directions by 1 pixel and 2 pixels. Also, the network can recognize all the distorted characters with 100% recognition rate. The simulations show that the test patterns cover a ${\pm}20^{\circ}$ range of rotation correctly. The proposed network can also recall correctly all the learned characters with 100% recognition rate. The proposed network is simple and its learning and recall speeds are very fast. This network also works for the segmentation and recognition of handwritten characters.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.10
• /
• pp.2643-2657
• /
• 2023

Scene graphs are structured representations that can clearly convey objects and the relationships between them, but are often heavily biased due to the highly skewed, long-tailed relational labeling in the dataset. Indeed, the visual world itself and its descriptions are biased. Therefore, Unbiased Scene Graph Generation (USGG) prefers to train models to eliminate long-tail effects as much as possible, rather than altering the dataset directly. To this end, we propose Geometric and Semantic Improvement (GSI) for USGG to mitigate this issue. First, to fully exploit the feature information in the images, geometric dimension and semantic dimension enhancement modules are designed. The geometric module is designed from the perspective that the position information between neighboring object pairs will affect each other, which can improve the recall rate of the overall relationship in the dataset. The semantic module further processes the embedded word vector, which can enhance the acquisition of semantic information. Then, to improve the recall rate of the tail data, the Class Balanced Seesaw Loss (CBSLoss) is designed for the tail data. The recall rate of the prediction is improved by penalizing the body or tail relations that are judged incorrectly in the dataset. The experimental findings demonstrate that the GSI method performs better than mainstream models in terms of the mean Recall@K (mR@K) metric in three tasks. The long-tailed imbalance in the Visual Genome 150 (VG150) dataset is addressed better using the GSI method than by most of the existing methods.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.681-684
• /
• 2021

In this paper, we study image clustering without labeling using machine learning techniques. We proposed an unsupervised machine learning technique to design an image clustering model that automatically categorizes images into groups. Our experiment focused on inception convolutional neural networks (inception V3) with k-mean methods to cluster images. For this, we collect the public datasets containing Food-K5, Flowers, Handwritten Digit, Cats-dogs, and our dataset Rice Germination, and the owner dataset Palm print. Our experiment can expand into three-part; First, format all the images to un-label and move to whole datasets. Second, load dataset into the inception V3 extraction image features and transferred to the k-mean cluster group hold on six classes. Lastly, evaluate modeling accuracy using the confusion matrix base on precision, recall, F1 to analyze. In this our methods, we can get the results as 1) Handwritten Digit (precision = 1.000, recall = 1.000, F1 = 1.00), 2) Food-K5 (precision = 0.975, recall = 0.945, F1 = 0.96), 3) Palm print (precision = 1.000, recall = 0.999, F1 = 1.00), 4) Cats-dogs (precision = 0.997, recall = 0.475, F1 = 0.64), 5) Flowers (precision = 0.610, recall = 0.982, F1 = 0.75), and our dataset 6) Rice Germination (precision = 0.997, recall = 0.943, F1 = 0.97). Our experiment showed that modeling could get an accuracy rate of 0.8908; the outcomes state that the proposed model is strongest enough to differentiate the different images and classify them into clusters.

• Korean Journal of Health Education and Promotion
• /
• v.24 no.4
• /
• pp.99-114
• /
• 2007

Object: To analyze the effect of a media-campaign for "walking exercise participation improvement", which impacted walking-related awareness and behavior change of residents in Seoul. Method: This study used three campaign media including printing information, walking exercise indication board and a public advertisement of cable TV to lead a walking-related awareness change and practice frequency(number of days per week walking) and time(minutes per day walking) of walking exercise. To evaluate the exposure and message-recall levels of a campaign and effects of awareness change and walking practice, this study used a questionnaire survey(N=377). Result: 1) Group of exposure to campaign more participate and had the higher frequency(p=.015) and time(p=.023) in walking exercise and in comparison with group of nonexposure. 2) Group of changed awareness to campaign more participate and had the higher frequency and time in walking exercise and in comparison with group of no changed perception(p <.05). 3) Level of message recall of ${\ulcorner}$printing information${\lrcorner}$ was associated with number of days per week walking, and level of message recall of ${\ulcorner}$public advertisement of cable TV${\lrcorner}$ was associated with minutes per day walking at a statistically significant level(p <.05). Conclusion: These results suggest that media campaign can enhance the success of community-based efforts to promote awareness change and walking practice.

• Journal of the Korean Data and Information Science Society
• /
• v.16 no.2
• /
• pp.263-270
• /
• 2005

The reliability guarantee insurance policy for parts and materials was introduced to the market in 2003. This policy indemnifies manufactures for the repair/failure costs, recall expenses. In this paper, owing to the nature of the policy, we propose a new rate-making system considering the type of product and industry, quality control circumstances, record of guarantee performance, and exposures.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.478-480
• /
• 2022

This paper studies human activity image classification using deep transfer learning techniques focused on the inception convolutional neural networks (InceptionV3) model. For this, we used UFC-101 public datasets containing a group of students' behaviors in mathematics classrooms at a school in Thailand. The video dataset contains Play Sitar, Tai Chi, Walking with Dog, and Student Study (our dataset) classes. The experiment was conducted in three phases. First, it extracts an image frame from the video, and a tag is labeled on the frame. Second, it loads the dataset into the inception V3 with transfer learning for image classification of four classes. Lastly, we evaluate the model's accuracy using precision, recall, F1-Score, and confusion matrix. The outcomes of the classifications for the public and our dataset are 1) Play Sitar (precision = 1.0, recall = 1.0, F1 = 1.0), 2), Tai Chi (precision = 1.0, recall = 1.0, F1 = 1.0), 3) Walking with Dog (precision = 1.0, recall = 1.0, F1 = 1.0), and 4) Student Study (precision = 1.0, recall = 1.0, F1 = 1.0), respectively. The results show that the overall accuracy of the classification rate is 100% which states the model is more powerful for learning UCF-101 and our dataset with higher accuracy.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.6
• /
• pp.175-186
• /
• 2004

We present a hybrid approach of texture-based method and connected component (CC)-based method for text extraction in complex images. Two primary methods, which are mainly utilized in this area, are sequentially merged for compensating for their weak points. An automatically constructed MLP-based texture classifier can increase recall rates for complex images with small amount of user intervention and without explicit feature extraction. CC-based filtering based on the shape information using NMF enhances the precision rate without affecting overall performance. As a result, a combination of texture and CC-based methods leads to not only robust but also efficient text extraction. We also enhance the processing speed by adopting appropriate region marking methods for each input image category.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.40 no.1
• /
• pp.19-29
• /
• 2003

This paper presents a novel hybrid classifier for face recognition with artificially generated virtual training samples. We utilize both the nearest neighbor approach in feature angle space and a connectionist model to obtain a synergy effect by combining the results of two heterogeneous classifiers. First, a classifier called the nearest feature angle (NFA), based on angular information, finds the most similar feature to the query from a given training set. Second, a classifier has been developed based on the recall of stored frontal projection of the query feature. It uses a frontal recall network (FRN) that finds the most similar frontal one among the stored frontal feature set. For FRN, we used an ensemble neural network consisting of multiple multiplayer perceptrons (MLPs), each of which is trained independently to enhance generalization capability. Further, both classifiers used the virtual training set generated adaptively, according to the spatial distribution of each person's training samples. Finally, the results of the two classifiers are combined to comprise the best matching class, and a corresponding similarit measure is used to make the final decision. The proposed classifier achieved an average classification rate of 96.33% against a large group of different test sets of images, and its average error rate is 61.5% that of the nearest feature line (NFL) method, and achieves a more robust classification performance.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.4
• /
• pp.1693-1697
• /
• 2014

Background: The Bahcesehir Breast Cancer Screening Project is the first organized population based breast cancer mammographic screening project in Turkey. The objective of this prospective observational study was to demonstrate the feasibility of a screening program in a developing country and to determine the appropriate age (40 or 50 years old) to start with screening in Turkish women. Materials and Methods: Between January 2009 to December 2010, a total of 3,758 women aged 40-69 years were recruited in this prospective study. Screening was conducted biannually, and five rounds were planned. After clinical breast examination (CBE), two-view mammograms were obtained. True positivity, false positivity, positive predictive values (PPV) according to ACR, cancer detection rate, minimal cancer detection rate, axillary node positivity and recall rate were calculated. Breast ultrasound and biopsy were performed in suspicious cases. Results: Breast biopsy was performed in 55 patients, and 18 cancers were detected in the first round. The overall cancer detection rate was 4.8 per 1,000 women. Most of the screened women (54%) and detected cancers (56%) were in women aged 40-49. Ductal carcinoma in situ (DCIS) and stage I cancer and axillary node positivity rates were 22%, 61%, and 16.6%, respectively. The positive predictivity for biopsy was 32.7%, whereas the overall recall rate was 18.4 %. Conclusions: Preliminary results of the study suggest that population based organized screening are feasible and age of onset of mammographic screening should be 40 years in Turkey.

• Restorative Dentistry and Endodontics
• /
• v.46 no.2
• /
• pp.29.1-29.11
• /
• 2021

Objectives: This study was conducted to compare the post-fracture survival rate of endodontically treated molar endodontically treated teeth (molar ETT) restored with resin composites or crowns and to identify potential risk factors, using a retrospective cohort design. Materials and Methods: Dental records of molar ETT with crowns or composite restorations (recall period, 2015-2019) were collected based on inclusion and exclusion criteria. The incidence of unrestorable fractures was identified, and molar ETT were classified according to survival. Information on potential risk factors was collected. Survival rates and potential risk factors were analyzed using the Kaplan-Meier log-rank test and Cox regression model. Results: The overall survival rate of molar ETT was 87% (mean recall period, 31.73 ± 17.56 months). The survival rates of molar ETT restored with composites and crowns were 81.6% and 92.7%, reflecting a significant difference (p < 0.05). However, ETT restored with composites showed a 100% survival rate if only 1 surface was lost, which was comparable to the survival rate of ETT with crowns. The survival rates of ETT with composites and crowns were significantly different (97.6% vs. 83.7%) in the short-term (12-24 months), but not in the long-term (> 24 months) (87.8% vs. 79.5%). Conclusions: The survival rate from fracture was higher for molar ETT restored with crowns was higher than for ETT restored with composites, especially in the first 2 years after restoration. Molar ETT with limited tooth structure loss only on the occlusal surface could be successfully restored with composite restorations.