• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.4
• /
• pp.95-105
• /
• 2021

Road surface damage detection is essential for a comfortable driving environment and the prevention of safety accidents. Road management institutes are using automated technology-based inspection equipment and systems. As one of these automation technologies, a sensor to detect road surface damage plays an important role. For this purpose, several studies on sensors using deep learning have been conducted in recent years. Road images and label images are needed to develop such deep learning algorithms. On the other hand, considerable time and labor will be needed to secure label images. In this paper, the adversarial learning method, one of the semi-supervised learning techniques, was proposed to solve this problem. For its implementation, a lightweight deep neural network model was trained using 5,327 road images and 1,327 label images. After experimenting with 400 road images, a model with a mean intersection over a union of 80.54% and an F1 score of 77.85% was developed. Through this, a technology that can improve recognition performance by adding only road images was developed to learning without label images and is expected to be used as a technology for road surface management in the future.

• Journal of the Korean Society of Industry Convergence
• /
• v.25 no.2_1
• /
• pp.177-184
• /
• 2022

The automatic defect sorting function of machinery parts is being introduced to the automation of the manufacturing process. In the final stage of automation of the manufacturing process, it is necessary to apply computer vision rather than human visual judgment to determine whether there is a defect. In this paper, we introduce a deep learning method to improve the classification performance of typical mechanical parts, such as welding parts, galvanized round plugs, and electro galvanized nuts, based on the results of experiments. In the case of poor welding, the method to further increase the depth of layer of the basic deep learning model was effective, and in the case of a circular plug, the surrounding data outside the defective target area affected it, so it could be solved through an appropriate pre-processing technique. Finally, in the case of a nut plated with zinc, since it receives data from multiple cameras due to its three-dimensional structure, it is greatly affected by lighting and has a problem in that it also affects the background image. To solve this problem, methods such as two-dimensional connectivity were applied in the object segmentation preprocessing process. Although the experiments suggested that the proposed methods are effective, most of the provided good/defective images data sets are relatively small, which may cause a learning balance problem of the deep learning model, so we plan to secure more data in the future.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.35 no.5
• /
• pp.259-266
• /
• 2022

In this study, we developed a deep learning and vision sensor-based assembly performance evaluation method isfor prefabricated steel structures. The assembly parts were segmented using a modified version of the receptive field block convolution module inspired by the eccentric function of the human visual system. The quality of the assembly was evaluated by detecting the bolt holes in the segmented assembly part and calculating the bolt hole positions. To validate the performance of the evaluation, models of standard and defective assembly parts were produced using a 3D printer. The assembly part segmentation network was trained based on the 3D model images captured from a vision sensor. The sbolt hole positions in the segmented assembly image were calculated using image processing techniques, and the assembly performance evaluation using the k-nearest neighbor algorithm was verified. The experimental results show that the assembly parts were segmented with high precision, and the assembly performance based on the positions of the bolt holes in the detected assembly part was evaluated with a classification error of less than 5%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.11
• /
• pp.3991-4010
• /
• 2021

The Question Similarity Measurement of Chinese Crop Diseases and Insect Pests (QSM-CCD&IP) aims to judge the user's tendency to ask questions regarding input problems. The measurement is the basis of the Agricultural Knowledge Question and Answering (Q & A) system, information retrieval, and other tasks. However, the corpus and measurement methods available in this field have some deficiencies. In addition, error propagation may occur when the word boundary features and local context information are ignored when the general method embeds sentences. Hence, these factors make the task challenging. To solve the above problems and tackle the Question Similarity Measurement task in this work, a corpus on Chinese crop diseases and insect pests(CCDIP), which contains 13 categories, was established. Then, taking the CCDIP as the research object, this study proposes a Chinese agricultural text similarity matching model, namely, the AgrCQS. This model is based on mixed information extraction. Specifically, the hybrid embedding layer can enrich character information and improve the recognition ability of the model on the word boundary. The multi-scale local information can be extracted by multi-core convolutional neural network based on multi-weight (MM-CNN). The self-attention mechanism can enhance the fusion ability of the model on global information. In this research, the performance of the AgrCQS on the CCDIP is verified, and three benchmark datasets, namely, AFQMC, LCQMC, and BQ, are used. The accuracy rates are 93.92%, 74.42%, 86.35%, and 83.05%, respectively, which are higher than that of baseline systems without using any external knowledge. Additionally, the proposed method module can be extracted separately and applied to other models, thus providing reference for related research.

• International Journal of Computer Science & Network Security
• /
• v.23 no.3
• /
• pp.151-160
• /
• 2023

The Hadith is the second source of Islamic jurisprudence after Qur'an. Both sources are indispensable for muslims to practice Islam. All Ahadith are collected and are written. But most books of Hadith contain Ahadith that can be weak or rejected. So, quite a long time, scholars of Hadith have defined laws, rules and principles of Hadith to know the correct Hadith (Sahih) from the fair (Hassen) and weak (Dhaif). Unfortunately, the application of these rules, laws and principles is done manually by the specialists or students until now. The work presented in this paper is part of the automatic treatment of Hadith, and more specifically, it aims to automatically process the chain of narrators (Hadith Isnad) to find its different components and affect for each component its own tag using a statistical method: the Hidden Markov Models (HMM). This method is a power abstraction for times series data and a robust tool for representing probability distributions over sequences of observations. In this paper, we describe an important tool in the Hadith isnad processing: A chunker with HMM. The role of this tool is to decompose the chain of narrators (Isnad) and determine the tag of each part of Isnad (POI). First, we have compiled a tagset containing 13 tags. Then, we have used these tags to manually conceive a corpus of 100 chains of narrators from "Sahih Alboukhari" and we have extracted a lexicon from this corpus. This lexicon is a set of XML documents based on HPSG features and it contains the information of 134 narrators. After that, we have designed and implemented an analyzer based on HMM that permit to assign for each part of Isnad its proper tag and for each narrator its features. The system was tested on 2661 not duplicated Isnad from "Sahih Alboukhari". The obtained result achieved F-scores of 93%.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.1
• /
• pp.91-97
• /
• 2009

This paper proposes novel FE-CBIRS that finds best position of a cut to be retrieved based on color feature distribution in digital contents of IPTV. Conventional CBIRS have used a method that utilizes both color and shape information together to classify images, as well as a method that utilizes both feature information of the entire region and feature information of a partial region that is extracted by segmentation for searching. Also, in the algorithm, average, standard deviation and skewness values are used in case of color features for each hue, saturation and intensity values respectively. Furthermore, in case of using partial regions, only a few major colors are used and in case of shape features, the invariant moment is mainly used on the extracted partial regions. Due to these reasons, some problems have been issued in CBIRS in processing time and accuracy so far. Therefore, in order to tackle these problems, this paper proposes the FE-CBIRS that makes searching speed faster by classifying and indexing the extracted color information by each class and by using several cuts that are restricted in range as comparative images.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.374-376
• /
• 2021

For the diagnosis of cancer-related diseases in clinical practice, pathological examination using biopsy is essential after basic diagnosis using imaging equipment. In order to proceed with such a biopsy, the assistance of an oncologist, clinical pathologist, etc. with specialized knowledge and the minimum required time are essential for confirmation. In recent years, research related to the establishment of a system capable of automatic classification of cancer cells using artificial intelligence is being actively conducted. However, previous studies show limitations in the type and accuracy of cells based on a limited algorithm. In this study, we propose a method to identify a total of 4 cancer cells through a convolutional neural network, a kind of deep learning. The optical images obtained through cell culture were learned through EfficientNet after performing pre-processing such as identification of the location of cells and image segmentation using OpenCV. The model used various hyper parameters based on EfficientNet, and trained InceptionV3 to compare and analyze the performance. As a result, cells were classified with a high accuracy of 96.8%, and this analysis method is expected to be helpful in confirming cancer.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.250-252
• /
• 2022

Recently, with the development of IoT technology and AI, unmanned and automated systems are progressing in various fields, and various application technologies are being studied in systems using algorithms such as object detection, recognition, and tracking. In the case of a system operating based on an image, noise removal is performed as a pre-processing process, and precise noise removal is sometimes required depending on the environment of the system. In this paper, we propose a modified central weight filter algorithm using pixel division of local regions to minimize the blurring that tends to occur in the filtering process and to emphasize the details of the resulting image. In the proposed algorithm, when a pixel of a local area is divided into two areas, the center of the dominant area among the divided areas is set as a criterion for the weight filter algorithm. The resulting image is calculated by convolving the transformed center weight with the pixel value inside the filtering mask.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.6-8
• /
• 2022

Many systems that use images through object recognition using deep learning have provided various solutions beyond the existing methods. Many studies have proven its usability, and the actual control system shows the possibility of using it to make people's work more convenient. Many studies have proven its usability, and actual control systems make human tasks more convenient and show possible. However, with hardware-intensive performance, the development of models is facing some limitations, and the ease with the use and additional utilization of many unupdated models is falling. In this paper, we propose how to increase utilization and accuracy by providing additional information on the emotional regions of colors and objects by utilizing learning and weights from HSV color histograms of local image data recognized after conventional stereotyped object recognition results.

• Journal of the Korean Society of Radiology
• /
• v.83 no.6
• /
• pp.1229-1239
• /
• 2022

Recently, artificial intelligence (AI) technology has shown potential clinical utility in a wide range of MRI fields. In particular, AI models for improving the efficiency of the image acquisition process and the quality of reconstructed images are being actively developed by the MR research community. AI is expected to further reduce acquisition times in various MRI protocols used in clinical practice when compared to current parallel imaging techniques. Additionally, AI can help with tasks such as planning, parameter optimization, artifact reduction, and quality assessment. Furthermore, AI is being actively applied to automate MR image analysis such as image registration, segmentation, and object detection. For this reason, it is important to consider the effects of protocols or devices in MR image analysis. In this review article, we briefly introduced issues related to AI application of MR image acquisition and reconstruction.