• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.5
• /
• pp.1039-1048
• /
• 2019

Symbolic execution, an automatic search method for vulnerability verification, has been technically improved over the last few years. However, it is still not practical to analyze the program using only the symbolic execution itself. One of the biggest reasons is that because of the path explosion problem that occurs during program analysis, there is not enough memory, and you can not find the solution of all paths in the program using symbolic execution. Thus, it is practical for the analyst to construct a path for symbolic execution to a target with vulnerability rather than solving all paths. In this paper, we propose a static analysis - based backward CFG(Control Flow Graph) generation technique that can be used in symbolic execution for program analysis. With the creation of a backward CFG, an analyst can select potential vulnerable points, and the backward path generated from that point can be used for future symbolic execution. We conducted experiments with Linux binaries(x86), and indeed showed that potential vulnerability selection and backward CFG path generation were possible in a variety of binary situations.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.11
• /
• pp.127-133
• /
• 2019

SQL is the most common language in data processing. Therefore, most of the colleges offer SQL in their curriculum. In this research, an auto scoring SQL test is proposed for the efficient results of SQL education. The system was treated with algorithms instead of using expensive DBMS(Data Base Management System) for automatic scoring, and satisfactory results were produced. For this system, the test question bank was established out of 'personnel management' and 'academic management'. It provides users with different sets of test each time. Scoring was done by dividing tables into two sections. The one that does not change the table(select) and the other that actually changes the table(update, insert, delete). In the case of a search, the answer and response were executed at first and then the results were compared and processed, the user's answers are evaluated by comparing the table with the correct answer. Modification, insertion, and deletion of table actually changes the data table, so data was restored by using ROLLBACK command. This system was implemented and tested 772 times on the 88 students in Computer Information Division of our college. The results of the implementation show that the average scoring time for a test consisting of 10 questions is 0.052 seconds, and the performance of this system is distinguished considering that multiple responses cannot be processed at the same time by a human grader, we want to develop a problem system that takes into account the difficulty of the problem into account near future.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.199-206
• /
• 2023

This study utilizes deep learning algorithms to automatically classify construction waste into three categories: wood waste, plastic waste, and concrete waste. Two models, VGG-16 and ViT (Vision Transformer), which are convolutional neural network image classification algorithms and NLP-based models that sequence images, respectively, were compared for their performance in classifying construction waste. Image data for construction waste was collected by crawling images from search engines worldwide, and 3,000 images, with 1,000 images for each category, were obtained by excluding images that were difficult to distinguish with the naked eye or that were duplicated and would interfere with the experiment. In addition, to improve the accuracy of the models, data augmentation was performed during training with a total of 30,000 images. Despite the unstructured nature of the collected image data, the experimental results showed that VGG-16 achieved an accuracy of 91.5%, and ViT achieved an accuracy of 92.7%. This seems to suggest the possibility of practical application in actual construction waste data management work. If object detection techniques or semantic segmentation techniques are utilized based on this study, more precise classification will be possible even within a single image, resulting in more accurate waste classification

• Journal of Internet Computing and Services
• /
• v.24 no.6
• /
• pp.127-135
• /
• 2023

With the development of information and communication technology, the amount of information produced and shared in the battlefield and stored and managed in the system dramatically increased. This means that the amount of information which cansupport situational awareness and decision making of the commanders has increased, but on the other hand, it is also a factor that hinders rapid decision making by increasing the information overload on the commanders. To overcome this limitation, this study proposes a method to automatically search, select, and summarize documents that can help the commanders to understand the battlefield situation reports that he or she received. First, named entities are discovered from the battlefield situation report using a named entity recognition method. Second, the documents related to each named entity are discovered. Third, a language model and collaborative filtering are used to select the documents. At this time, the language model is used to calculate the similarity between the received report and the discovered documents, and collaborative filtering is used to reflect the commander's document reading history. Finally, sentences containing each named entity are selected from the documents and sorted. The experiment was carried out using academic papers since their characteristics are similar to military documents, and the validity of the proposed method was verified.

• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.1-17
• /
• 2017

The deep learning framework is software designed to help develop deep learning models. Some of its important functions include "automatic differentiation" and "utilization of GPU". The list of popular deep learning framework includes Caffe (BVLC) and Theano (University of Montreal). And recently, Microsoft's deep learning framework, Microsoft Cognitive Toolkit, was released as open-source license, following Google's Tensorflow a year earlier. The early deep learning frameworks have been developed mainly for research at universities. Beginning with the inception of Tensorflow, however, it seems that companies such as Microsoft and Facebook have started to join the competition of framework development. Given the trend, Google and other companies are expected to continue investing in the deep learning framework to bring forward the initiative in the artificial intelligence business. From this point of view, we think it is a good time to compare some of deep learning frameworks. So we compare three deep learning frameworks which can be used as a Python library. Those are Google's Tensorflow, Microsoft's CNTK, and Theano which is sort of a predecessor of the preceding two. The most common and important function of deep learning frameworks is the ability to perform automatic differentiation. Basically all the mathematical expressions of deep learning models can be represented as computational graphs, which consist of nodes and edges. Partial derivatives on each edge of a computational graph can then be obtained. With the partial derivatives, we can let software compute differentiation of any node with respect to any variable by utilizing chain rule of Calculus. First of all, the convenience of coding is in the order of CNTK, Tensorflow, and Theano. The criterion is simply based on the lengths of the codes and the learning curve and the ease of coding are not the main concern. According to the criteria, Theano was the most difficult to implement with, and CNTK and Tensorflow were somewhat easier. With Tensorflow, we need to define weight variables and biases explicitly. The reason that CNTK and Tensorflow are easier to implement with is that those frameworks provide us with more abstraction than Theano. We, however, need to mention that low-level coding is not always bad. It gives us flexibility of coding. With the low-level coding such as in Theano, we can implement and test any new deep learning models or any new search methods that we can think of. The assessment of the execution speed of each framework is that there is not meaningful difference. According to the experiment, execution speeds of Theano and Tensorflow are very similar, although the experiment was limited to a CNN model. In the case of CNTK, the experimental environment was not maintained as the same. The code written in CNTK has to be run in PC environment without GPU where codes execute as much as 50 times slower than with GPU. But we concluded that the difference of execution speed was within the range of variation caused by the different hardware setup. In this study, we compared three types of deep learning framework: Theano, Tensorflow, and CNTK. According to Wikipedia, there are 12 available deep learning frameworks. And 15 different attributes differentiate each framework. Some of the important attributes would include interface language (Python, C ++, Java, etc.) and the availability of libraries on various deep learning models such as CNN, RNN, DBN, and etc. And if a user implements a large scale deep learning model, it will also be important to support multiple GPU or multiple servers. Also, if you are learning the deep learning model, it would also be important if there are enough examples and references.

• Journal of Intelligence and Information Systems
• /
• v.24 no.3
• /
• pp.1-19
• /
• 2018

Large amount of data is now available for research and business sectors to extract knowledge from it. This data can be in the form of unstructured data such as audio, text, and image data and can be analyzed by deep learning methodology. Deep learning is now widely used for various estimation, classification, and prediction problems. Especially, fashion business adopts deep learning techniques for apparel recognition, apparel search and retrieval engine, and automatic product recommendation. The core model of these applications is the image classification using Convolutional Neural Networks (CNN). CNN is made up of neurons which learn parameters such as weights while inputs come through and reach outputs. CNN has layer structure which is best suited for image classification as it is comprised of convolutional layer for generating feature maps, pooling layer for reducing the dimensionality of feature maps, and fully-connected layer for classifying the extracted features. However, most of the classification models have been trained using online product image, which is taken under controlled situation such as apparel image itself or professional model wearing apparel. This image may not be an effective way to train the classification model considering the situation when one might want to classify street fashion image or walking image, which is taken in uncontrolled situation and involves people's movement and unexpected pose. Therefore, we propose to train the model with runway apparel image dataset which captures mobility. This will allow the classification model to be trained with far more variable data and enhance the adaptation with diverse query image. To achieve both convergence and generalization of the model, we apply Transfer Learning on our training network. As Transfer Learning in CNN is composed of pre-training and fine-tuning stages, we divide the training step into two. First, we pre-train our architecture with large-scale dataset, ImageNet dataset, which consists of 1.2 million images with 1000 categories including animals, plants, activities, materials, instrumentations, scenes, and foods. We use GoogLeNet for our main architecture as it has achieved great accuracy with efficiency in ImageNet Large Scale Visual Recognition Challenge (ILSVRC). Second, we fine-tune the network with our own runway image dataset. For the runway image dataset, we could not find any previously and publicly made dataset, so we collect the dataset from Google Image Search attaining 2426 images of 32 major fashion brands including Anna Molinari, Balenciaga, Balmain, Brioni, Burberry, Celine, Chanel, Chloe, Christian Dior, Cividini, Dolce and Gabbana, Emilio Pucci, Ermenegildo, Fendi, Giuliana Teso, Gucci, Issey Miyake, Kenzo, Leonard, Louis Vuitton, Marc Jacobs, Marni, Max Mara, Missoni, Moschino, Ralph Lauren, Roberto Cavalli, Sonia Rykiel, Stella McCartney, Valentino, Versace, and Yve Saint Laurent. We perform 10-folded experiments to consider the random generation of training data, and our proposed model has achieved accuracy of 67.2% on final test. Our research suggests several advantages over previous related studies as to our best knowledge, there haven't been any previous studies which trained the network for apparel image classification based on runway image dataset. We suggest the idea of training model with image capturing all the possible postures, which is denoted as mobility, by using our own runway apparel image dataset. Moreover, by applying Transfer Learning and using checkpoint and parameters provided by Tensorflow Slim, we could save time spent on training the classification model as taking 6 minutes per experiment to train the classifier. This model can be used in many business applications where the query image can be runway image, product image, or street fashion image. To be specific, runway query image can be used for mobile application service during fashion week to facilitate brand search, street style query image can be classified during fashion editorial task to classify and label the brand or style, and website query image can be processed by e-commerce multi-complex service providing item information or recommending similar item.

• Journal of dental hygiene science
• /
• v.10 no.3
• /
• pp.177-183
• /
• 2010

This study was begun to search effect of contact angles of elastic rubber impression materials on the surface of working cast. Of elastic rubber impression materials with a Type III consistency, such as polysulfide, polyether and addition silicone, we selected one and then measured the contact angle after dripping a distilled water 3.3ml. Then, after pouring a dental anhydrite in three types of impression materials, we prepared a working cast and then examined its surface. Contact angle was measured using a full automatic contact angle measuring system (DM-700, KYOWA, Japan), and the surface of working cast was observed using a field emission scanning electron microscope (JSM-6700F, JEOL Ltd., JAPAN). The following results were obtained: 1) $Mean{\pm}SD$ (SD: standard deviation) of the initial contact angles were $91.3{\pm}20.5^{\circ}$ in the addition silicone materials, $90.0{\pm}2.2^{\circ}$ in the polyethers and $101.5{\pm}2.3^{\circ}$ in the polysulfides. These results indicate that mean values were similar but standard deviations of the three materials showed a great discrepancy. 2) As the time elapsed, addition silicone materials were found to have a contact angle decreased abruptly as compared with the remaining two types. That is, the initial contact angle was $91.3^{\circ}$ and it was abruptly decreased to $29.4^{\circ}$ after 25 seconds. 3) In the polyethers, the initial contact angle was $101.5^{\circ}$ and it was decreased to $90.7^{\circ}$ after 25 seconds. In the polysulfides, however, the initial contact angle was $90.0^{\circ}$ and it was $84.2^{\circ}$ after 25 seconds. This showed almost no changes in the initial contact angles. Moreover, its magnitude was greater than that seen in additional silicones. 4) There were significant differences in the contact angles between the three types of elastic rubber impression materials as the time elapsed (p<0.001). On an observation on the surface of working cast, addition silicone materials were found to have the most dense surface. This was followed by polysulfides and polyethers in a descending order.

• Journal of Navigation and Port Research
• /
• v.39 no.1
• /
• pp.45-53
• /
• 2015

These days Container Terminals are focusing on increasing the quantity of containers and shipping lines choose Terminals by referring to the key elements of a terminal to perform the overall operation the fastest such as the location of the terminal, discharging ability, keeping environment, and other elements related to shipping in general. Container terminal is able to offer On-Dock service has become an important factor for shipping lines to choose that terminal. In this paper, we propose an algorithm for On-Dock system work algorithm, the algorithm Empty container exports, Full Container algorithm and The aim of our study focus on both container's gate out time and search for the effective terminal operation which is using the general On-Dock system through several algorithm like container batch priority, gate in and out job priority and empty container yard equipment allocation rule based on the automatic allocation method and manual allocation scheme for container. Gathering these information, it gives the priority and yard location of gate-out containers to control. That is, by selecting an optimum algorithm container, container terminals Empty reduces the container taken out time, it is possible to minimize unnecessary re-handling of the yard container can be enhanced with respect to the efficiency of the equipment. Operations and operating results of the Non On-Dock and On-Dock system is operated by the out work operations (scenarios) forms that are operating in the real Gwangyang Container Terminal derived results. Gwangyang Container terminal and apply the On-Dock system, Non On-Dock can be taken out this time, about 5 minutes more quickly when applying the system. when managing export orders for berths where On-Dock service is needed, ball containers are allocated and for import cargoes, D/O is managed and after carryout, return management, container damage, cleaning, fixing and controlling services are supported hence the berth service can be strengthened and container terminal business can grow.