• The Journal of the Korea Contents Association
• /
• v.21 no.6
• /
• pp.701-715
• /
• 2021

The drastic advance of recent deep learning technologies is heavily dependent on training datasets which are essential to train models by themselves with less human efforts. In comparison with the work to design deep learning models, preparing datasets is a long haul; at the moment, in the domain of vision intelligent, datasets are still being made by handwork requiring a lot of time and efforts, where workers need to directly make labels on each image usually with GUI-based labeling tools. In this paper, we overview the current status of vision datasets focusing on what datasets are being shared and how they are prepared with various labeling tools. Particularly, in order to relieve the repetitive and tiring labeling work, we present an interactive smart image annotating system with which the annotation work can be transformed from the direct human-only manual labeling to a correction-after-checking by means of a support of automatic labeling. In an experiment, we show that automatic labeling can greatly improve the productivity of datasets especially reducing time and efforts to specify regions of objects found in images. Finally, we discuss critical issues that we faced in the experiment to our annotation system and describe future work to raise the productivity of image datasets creation for accelerating AI technology.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.79-104
• /
• 2020

Recently, as deep learning has attracted attention, the use of deep learning is being considered as a method for solving problems in various fields. In particular, deep learning is known to have excellent performance when applied to applying unstructured data such as text, sound and images, and many studies have proven its effectiveness. Owing to the remarkable development of text and image deep learning technology, interests in image captioning technology and its application is rapidly increasing. Image captioning is a technique that automatically generates relevant captions for a given image by handling both image comprehension and text generation simultaneously. In spite of the high entry barrier of image captioning that analysts should be able to process both image and text data, image captioning has established itself as one of the key fields in the A.I. research owing to its various applicability. In addition, many researches have been conducted to improve the performance of image captioning in various aspects. Recent researches attempt to create advanced captions that can not only describe an image accurately, but also convey the information contained in the image more sophisticatedly. Despite many recent efforts to improve the performance of image captioning, it is difficult to find any researches to interpret images from the perspective of domain experts in each field not from the perspective of the general public. Even for the same image, the part of interests may differ according to the professional field of the person who has encountered the image. Moreover, the way of interpreting and expressing the image also differs according to the level of expertise. The public tends to recognize the image from a holistic and general perspective, that is, from the perspective of identifying the image's constituent objects and their relationships. On the contrary, the domain experts tend to recognize the image by focusing on some specific elements necessary to interpret the given image based on their expertise. It implies that meaningful parts of an image are mutually different depending on viewers' perspective even for the same image. So, image captioning needs to implement this phenomenon. Therefore, in this study, we propose a method to generate captions specialized in each domain for the image by utilizing the expertise of experts in the corresponding domain. Specifically, after performing pre-training on a large amount of general data, the expertise in the field is transplanted through transfer-learning with a small amount of expertise data. However, simple adaption of transfer learning using expertise data may invoke another type of problems. Simultaneous learning with captions of various characteristics may invoke so-called 'inter-observation interference' problem, which make it difficult to perform pure learning of each characteristic point of view. For learning with vast amount of data, most of this interference is self-purified and has little impact on learning results. On the contrary, in the case of fine-tuning where learning is performed on a small amount of data, the impact of such interference on learning can be relatively large. To solve this problem, therefore, we propose a novel 'Character-Independent Transfer-learning' that performs transfer learning independently for each character. In order to confirm the feasibility of the proposed methodology, we performed experiments utilizing the results of pre-training on MSCOCO dataset which is comprised of 120,000 images and about 600,000 general captions. Additionally, according to the advice of an art therapist, about 300 pairs of 'image / expertise captions' were created, and the data was used for the experiments of expertise transplantation. As a result of the experiment, it was confirmed that the caption generated according to the proposed methodology generates captions from the perspective of implanted expertise whereas the caption generated through learning on general data contains a number of contents irrelevant to expertise interpretation. In this paper, we propose a novel approach of specialized image interpretation. To achieve this goal, we present a method to use transfer learning and generate captions specialized in the specific domain. In the future, by applying the proposed methodology to expertise transplant in various fields, we expected that many researches will be actively conducted to solve the problem of lack of expertise data and to improve performance of image captioning.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.2
• /
• pp.631-638
• /
• 2017

In the age of the Fourth Industrial Revolution, lifestyle and industrial structures are faced with evolution from IT-based automation to the intelligent stage, demanding talents with software capabilities in various fields. Reflecting these demands, the government has enhanced basic software education for non-majors in elementary and secondary schools as well as universities. In this study, the software convergence education of Non-Majors is proposed to improve the general problem solving ability based on computational thinking and the software convergence ability in the field of their own by developing robot activity. The subjects of this study were 91 students, who were composed of various majors. The class was designed with computing thinking, convergence elements, and creative robot activity. The study was conducted for 13 weeks. To examine the effects of software convergence education through the creative robot activity, this study observed changes in the students' learning outcomes, satisfaction with creative robot activities, and perceptions of other disciplines after class based on pre-diagnosis surveys. The survey asked 12 questions including an understanding of the learning contents, overall satisfaction with multidisciplinary collaborative learning, understanding of other disciplines, and self-evaluation of problem solving ability through creative robot activities, which were compared with that before the class. They answered that their ability was improved.

• Journal of Intelligence and Information Systems
• /
• v.12 no.3
• /
• pp.67-82
• /
• 2006

We propose the HBEM(Hybrid Behavior Evolution Model) composed of rule classification and evolutionary neural network using rule descriptor and link descriptor for evolutionary behavior of virtual robots. In our model, two levels of the knowledge of behaviors were represented. In the upper level, the representation was improved using rule and link descriptors together. And then in the lower level, behavior knowledge was represented in form of bit string and learned adapting their chromosomes by the genetic operators. A virtual robot was composed by the learned chromosome which had the best fitness. The composed virtual robot perceives the surrounding situations and they were classifying the pattern through rules and processing the result in neural network and behaving. To evaluate our proposed model, we developed HBES(Hybrid Behavior Evolution System) and adapted the problem of gathering food of the virtual robots. In the results of testing our system, the learning time was fewer than the evolution neural network of the condition which was same. And then, to evaluate the effect improving the fitness by the rules we respectively measured the fitness adapted or not about the chromosomes where the learning was completed. In the results of evaluating, if the rules were not adapted the fitness was lowered. It showed that our proposed model was better in the learning performance and more regular than the evolutionary neural network in the behavior evolution of the virtual robots.

• Journal of Broadcast Engineering
• /
• v.17 no.2
• /
• pp.270-283
• /
• 2012

With the advent of multi-channel TV, IPTV and smart TV services, excessive amounts of TV program contents become available at users' sides, which makes it very difficult for TV viewers to easily find and consume their preferred TV programs. Therefore, the service of automatic TV recommendation is an important issue for TV users for future intelligent TV services, which allows to improve access to their preferred TV contents. In this paper, we present a recommendation model based on statistical machine learning using a collaborative filtering concept by taking in account both public and personal preferences on TV program contents. For this, users' preference on TV programs is modeled as a latent topic variable using LDA (Latent Dirichlet Allocation) which is recently applied in various application domains. To apply LDA for TV recommendation appropriately, TV viewers's interested topics is regarded as latent topics in LDA, and asymmetric Dirichlet distribution is applied on the LDA which can reveal the diversity of the TV viewers' interests on topics based on the analysis of the real TV usage history data. The experimental results show that the proposed LDA based TV recommendation method yields average 66.5% with top 5 ranked TV programs in weekly recommendation, average 77.9% precision in bimonthly recommendation with top 5 ranked TV programs for the TV usage history data of similar taste user groups.

• Journal of the Korea Convergence Society
• /
• v.10 no.11
• /
• pp.55-61
• /
• 2019

In order to realize foreign language education, various existing educational media have been provided, but there are disadvantages in that the cost of the parish and the media program is high and the real-time responsiveness is poor. In this paper, we propose an artificial intelligence English conversation system based on VR and speech recognition. We used Google CardBoard VR and Google Speech API to build the system and developed artificial intelligence algorithms for providing virtual reality environment and talking. In the proposed speech recognition server system, the sentences spoken by the user can be divided into word units and compared with the data words stored in the database to provide the highest probability. Users can communicate with and respond to people in virtual reality. The function provided by the conversation is independent of the contextual conversations and themes, and the conversations with the AI assistant are implemented in real time so that the user system can be checked in real time. It is expected to contribute to the expansion of virtual education contents service related to the Fourth Industrial Revolution through the system combining the virtual reality and the voice recognition function proposed in this paper.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.6
• /
• pp.1317-1322
• /
• 2021

Software in the era of the Fourth Industrial Revolution is becoming a key foundation in an intelligent information society. Therefore, it is necessary to study the new direction of manpower training and education that can cope with the times. To this end, the Ministry of Education reorganized the curriculum and is implementing software education based on a logical problem-solving process based on computing thinking skills rather than acquiring general ICT knowledge. However, there is a lack of securing high-quality educational content for software education, and there is also a lack of teaching aids that can be taught in connection with advanced IT technologies. To overcome this, this paper proposes the development of external expansion devices to expand educational content and functions capable of convergent software education such as artificial intelligence using coding robots for software education. Through this, effective software education is possible by improving the curriculum of the existing simple problem-solving method and developing various learning materials.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.44-55
• /
• 2022

In order to automatically provide information on fruits in agricultural related broadcasting contents, instance image segmentation of target fruits is required. In addition, the information on the 3D pose of the corresponding fruit may be meaningfully used. This paper represents research that provides information about tomatoes in video content. A large amount of data is required to learn the instance segmentation, but it is difficult to obtain sufficient training data. Therefore, the training data is generated through a data augmentation technique based on a small amount of real images. Compared to the result using only the real images, it is shown that the detection performance is improved as a result of learning through the synthesized image created by separating the foreground and background. As a result of learning augmented images using images created using conventional image pre-processing techniques, it was shown that higher performance was obtained than synthetic images in which foreground and background were separated. To estimate the pose from the result of object detection, a point cloud was obtained using an RGB-D camera. Then, cylinder fitting based on least square minimization is performed, and the tomato pose is estimated through the axial direction of the cylinder. We show that the results of detection, instance image segmentation, and cylinder fitting of a target object effectively through various experiments.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.141-166
• /
• 2019

Recently, channels like social media and SNS create enormous amount of data. In all kinds of data, portions of unstructured data which represented as text data has increased geometrically. But there are some difficulties to check all text data, so it is important to access those data rapidly and grasp key points of text. Due to needs of efficient understanding, many studies about text summarization for handling and using tremendous amounts of text data have been proposed. Especially, a lot of summarization methods using machine learning and artificial intelligence algorithms have been proposed lately to generate summary objectively and effectively which called "automatic summarization". However almost text summarization methods proposed up to date construct summary focused on frequency of contents in original documents. Those summaries have a limitation for contain small-weight subjects that mentioned less in original text. If summaries include contents with only major subject, bias occurs and it causes loss of information so that it is hard to ascertain every subject documents have. To avoid those bias, it is possible to summarize in point of balance between topics document have so all subject in document can be ascertained, but still unbalance of distribution between those subjects remains. To retain balance of subjects in summary, it is necessary to consider proportion of every subject documents originally have and also allocate the portion of subjects equally so that even sentences of minor subjects can be included in summary sufficiently. In this study, we propose "subject-balanced" text summarization method that procure balance between all subjects and minimize omission of low-frequency subjects. For subject-balanced summary, we use two concept of summary evaluation metrics "completeness" and "succinctness". Completeness is the feature that summary should include contents of original documents fully and succinctness means summary has minimum duplication with contents in itself. Proposed method has 3-phases for summarization. First phase is constructing subject term dictionaries. Topic modeling is used for calculating topic-term weight which indicates degrees that each terms are related to each topic. From derived weight, it is possible to figure out highly related terms for every topic and subjects of documents can be found from various topic composed similar meaning terms. And then, few terms are selected which represent subject well. In this method, it is called "seed terms". However, those terms are too small to explain each subject enough, so sufficient similar terms with seed terms are needed for well-constructed subject dictionary. Word2Vec is used for word expansion, finds similar terms with seed terms. Word vectors are created after Word2Vec modeling, and from those vectors, similarity between all terms can be derived by using cosine-similarity. Higher cosine similarity between two terms calculated, higher relationship between two terms defined. So terms that have high similarity values with seed terms for each subjects are selected and filtering those expanded terms subject dictionary is finally constructed. Next phase is allocating subjects to every sentences which original documents have. To grasp contents of all sentences first, frequency analysis is conducted with specific terms that subject dictionaries compose. TF-IDF weight of each subjects are calculated after frequency analysis, and it is possible to figure out how much sentences are explaining about each subjects. However, TF-IDF weight has limitation that the weight can be increased infinitely, so by normalizing TF-IDF weights for every subject sentences have, all values are changed to 0 to 1 values. Then allocating subject for every sentences with maximum TF-IDF weight between all subjects, sentence group are constructed for each subjects finally. Last phase is summary generation parts. Sen2Vec is used to figure out similarity between subject-sentences, and similarity matrix can be formed. By repetitive sentences selecting, it is possible to generate summary that include contents of original documents fully and minimize duplication in summary itself. For evaluation of proposed method, 50,000 reviews of TripAdvisor are used for constructing subject dictionaries and 23,087 reviews are used for generating summary. Also comparison between proposed method summary and frequency-based summary is performed and as a result, it is verified that summary from proposed method can retain balance of all subject more which documents originally have.

• Journal of Intelligence and Information Systems
• /
• v.18 no.3
• /
• pp.53-77
• /
• 2012

This study analyses the difference of contents and tones of arguments among three Korean major newspapers, the Kyunghyang Shinmoon, the HanKyoreh, and the Dong-A Ilbo. It is commonly accepted that newspapers in Korea explicitly deliver their own tone of arguments when they talk about some sensitive issues and topics. It could be controversial if readers of newspapers read the news without being aware of the type of tones of arguments because the contents and the tones of arguments can affect readers easily. Thus it is very desirable to have a new tool that can inform the readers of what tone of argument a newspaper has. This study presents the results of clustering and classification techniques as part of text mining analysis. We focus on six main subjects such as Culture, Politics, International, Editorial-opinion, Eco-business and National issues in newspapers, and attempt to identify differences and similarities among the newspapers. The basic unit of text mining analysis is a paragraph of news articles. This study uses a keyword-network analysis tool and visualizes relationships among keywords to make it easier to see the differences. Newspaper articles were gathered from KINDS, the Korean integrated news database system. KINDS preserves news articles of the Kyunghyang Shinmun, the HanKyoreh and the Dong-A Ilbo and these are open to the public. This study used these three Korean major newspapers from KINDS. About 3,030 articles from 2008 to 2012 were used. International, national issues and politics sections were gathered with some specific issues. The International section was collected with the keyword of 'Nuclear weapon of North Korea.' The National issues section was collected with the keyword of '4-major-river.' The Politics section was collected with the keyword of 'Tonghap-Jinbo Dang.' All of the articles from April 2012 to May 2012 of Eco-business, Culture and Editorial-opinion sections were also collected. All of the collected data were handled and edited into paragraphs. We got rid of stop-words using the Lucene Korean Module. We calculated keyword co-occurrence counts from the paired co-occurrence list of keywords in a paragraph. We made a co-occurrence matrix from the list. Once the co-occurrence matrix was built, we used the Cosine coefficient matrix as input for PFNet(Pathfinder Network). In order to analyze these three newspapers and find out the significant keywords in each paper, we analyzed the list of 10 highest frequency keywords and keyword-networks of 20 highest ranking frequency keywords to closely examine the relationships and show the detailed network map among keywords. We used NodeXL software to visualize the PFNet. After drawing all the networks, we compared the results with the classification results. Classification was firstly handled to identify how the tone of argument of a newspaper is different from others. Then, to analyze tones of arguments, all the paragraphs were divided into two types of tones, Positive tone and Negative tone. To identify and classify all of the tones of paragraphs and articles we had collected, supervised learning technique was used. The Na$\ddot{i}$ve Bayesian classifier algorithm provided in the MALLET package was used to classify all the paragraphs in articles. After classification, Precision, Recall and F-value were used to evaluate the results of classification. Based on the results of this study, three subjects such as Culture, Eco-business and Politics showed some differences in contents and tones of arguments among these three newspapers. In addition, for the National issues, tones of arguments on 4-major-rivers project were different from each other. It seems three newspapers have their own specific tone of argument in those sections. And keyword-networks showed different shapes with each other in the same period in the same section. It means that frequently appeared keywords in articles are different and their contents are comprised with different keywords. And the Positive-Negative classification showed the possibility of classifying newspapers' tones of arguments compared to others. These results indicate that the approach in this study is promising to be extended as a new tool to identify the different tones of arguments of newspapers.