• Journal of the Korea Society of Computer and Information
• /
• v.26 no.2
• /
• pp.19-25
• /
• 2021

In this paper proposes a method of finding missing persons based on face-recognition technology and deep learning. In this paper, a real-time face-recognition technology was developed, which performs face verification and improves the accuracy of face identification through data fortification for face recognition and convolutional neural network(CNN)-based image learning after the pre-processing of images transmitted from a mobile device. In identifying a missing person's image using the system implemented in this paper, the model that learned both original and blur-processed data performed the best. Further, a model using the pre-learned Noisy Student outperformed the one not using the same, but it has had a limitation of producing high levels of deflection and dispersion.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.1
• /
• pp.51-73
• /
• 2023

Social media play a significant role in communicating information across the globe, connecting with loved ones, getting the news, communicating ideas, etc. However, a group of people uses social media to spread fake information, which has a bad impact on society. Therefore, minimizing fake news and its detection are the two primary challenges that need to be addressed. This paper presents a multi-modal deep learning technique to address the above challenges. The proposed modal can use and process visual and textual features. Therefore, it has the ability to detect fake information from visual and textual data. We used EfficientNetB0 and a sentence transformer, respectively, for detecting counterfeit images and for textural learning. Feature embedding is performed at individual channels, whilst fusion is done at the last classification layer. The late fusion is applied intentionally to mitigate the noisy data that are generated by multi-modalities. Extensive experiments are conducted, and performance is evaluated against state-of-the-art methods. Three real-world benchmark datasets, such as MediaEval (Twitter), Weibo, and Fakeddit, are used for experimentation. Result reveals that the proposed modal outperformed the state-of-the-art methods and achieved an accuracy of 86.48%, 82.50%, and 88.80%, respectively, for MediaEval (Twitter), Weibo, and Fakeddit datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.10
• /
• pp.2788-2808
• /
• 2023

Brain tumors are one of the most threatening malignancies for humans. Misdiagnosis of brain tumors can result in false medical intervention, which ultimately reduces a patient's chance of survival. Manual identification and segmentation of brain tumors from Magnetic Resonance Imaging (MRI) scans can be difficult and error-prone because of the great range of tumor tissues that exist in various individuals and the similarity of normal tissues. To overcome this limitation, the Amended Convolutional Neural Network (ACNN) model has been introduced, a unique combination of three techniques that have not been previously explored for brain tumor detection. The three techniques integrated into the ACNN model are image tissue preprocessing using the Kalman Bucy Smoothing Filter to remove noisy pixels from the input, image tissue segmentation using the Isotonic Regressive Image Tissue Segmentation Process, and feature extraction using the Marr Wavelet Transformation. The extracted features are compared with the testing features using a sigmoid activation function in the output layer. The experimental findings show that the suggested model outperforms existing techniques concerning accuracy, precision, sensitivity, dice score, Jaccard index, specificity, Positive Predictive Value, Hausdorff distance, recall, and F1 score. The proposed ACNN model achieved a maximum accuracy of 98.8%, which is higher than other existing models, according to the experimental results.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.5
• /
• pp.909-916
• /
• 2009

This paper presents the automatic vision algorithm to generate and calibrate reference phase plane to improve the accuracy of 3D measuring machine of using phase shifting projection moire method, which is not traditional N-bucket method, but is based on direct image processing method to the pattern projection image. Generally, to acquire accurate reference phase plane, the calibration specimen with well treated surface is needed, and detailed calibration method should be performed. For the cost reduction of specimen manufacturing and the calibration time reduction, on the specimen, not specially designed, with general accuracy level, an efficient calibration procedure for the reference phase generation is proposed. The proposed vision algorithm is developed to extract the line center points of the projected line pattern from acquired images, derive the line feature information consisting of its slope and intercept by using sampled feature points, and finally generate the related reference phase between line pairs. Experimental results show that the proposed method make reference phase plane with a good accuracy under noisy environment and the proposed algorithm can reduce the total cost to make high accurate calibration specimen, also increase the accuracy of reference phase plane, and reduce the complex calibration procedure to move grid via N-bucket algorithm precisely.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.3
• /
• pp.164-173
• /
• 2014

In underwater robotics, vision would be a key element for recognition in underwater environments. However, due to turbidity an underwater optical camera is rarely available. An underwater imaging sonar, as an alternative, delivers low quality sonar images which are not stable and accurate enough to find out natural objects by image processing. For this, artificial landmarks based on the characteristics of ultrasonic waves and their recognition method by a shape matrix transformation were proposed and were proven in Part 1. But, this is not working properly in undulating and dynamically noisy sea-bottom. To solve this, we propose a framework providing a selection phase of likelihood candidates, a selection phase for final candidates, recognition phase and tracking phase in sequence images, where a particle filter based selection mechanism to eliminate fake candidates and a mean shift based tracking algorithm are also proposed. All 4 steps are running in parallel and real-time processing. The proposed framework is flexible to add and to modify internal algorithms. A pool test and sea trial are carried out to prove the performance, and detail analysis of experimental results are done. Information is obtained from tracking phase such as relative distance, bearing will be expected to be used for control and navigation of underwater robots.

• Convergence Security Journal
• /
• v.19 no.2
• /
• pp.91-104
• /
• 2019

In this paper, we present a novel local descriptor, Local Prominent Directional Pattern (LPDP), to represent the description of facial images for gender recognition purpose. To achieve a clearly discriminative representation of local shape, presented method encodes a target pixel with the prominent directional variations in local structure from an analysis of statistics encompassed in the histogram of such directional variations. Use of the statistical information comes from the observation that a local neighboring region, having an edge going through it, demonstrate similar gradient directions, and hence, the prominent accumulations, accumulated from such gradient directions provide a solid base to represent the shape of that local structure. Unlike the sole use of gradient direction of a target pixel in existing methods, our coding scheme selects prominent edge directions accumulated from more samples (e.g., surrounding neighboring pixels), which, in turn, minimizes the effect of noise by suppressing the noisy accumulations of single or fewer samples. In this way, the presented encoding strategy provides the more discriminative shape of local structures while ensuring robustness to subtle changes such as local noise. We conduct extensive experiments on gender recognition datasets containing a wide range of challenges such as illumination, expression, age, and pose variations as well as sketch images, and observe the better performance of LPDP descriptor against existing local descriptors.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_3
• /
• pp.1761-1775
• /
• 2022

Due to unpredictable climate change, the frequency of occurrence of water-related disasters and the scale of damage are also continuously increasing. In terms of disaster management, it is essential to identify the damaged area in a wide area and monitor for mid-term and long-term forecasting. In the field of water disasters, research on remote sensing technology using Synthetic Aperture Radar (SAR) satellite images for wide-area monitoring is being actively conducted. Time-series analysis for monitoring requires a complex preprocessing process that collects a large amount of images and considers the noisy radar characteristics, and for this, a considerable amount of time is required. With the recent development of cloud computing technology, many platforms capable of performing spatiotemporal analysis using satellite big data have been proposed. Google Earth Engine (GEE)is a representative platform that provides about 600 satellite data for free and enables semi real time space time analysis based on the analysis preparation data of satellite images. Therefore, in this study, immediate water disaster damage detection and mid to long term time series observation studies were conducted using GEE. Through the Otsu technique, which is mainly used for change detection, changes in river width and flood area due to river flooding were confirmed, centered on the torrential rains that occurred in 2020. In addition, in terms of disaster management, the change trend of the time series waterbody from 2018 to 2022 was confirmed. The short processing time through javascript based coding, and the strength of spatiotemporal analysis and result expression, are expected to enable use in the field of water disasters. In addition, it is expected that the field of application will be expanded through connection with various satellite bigdata in the future.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.2
• /
• pp.227-232
• /
• 2018

In a Hough transform that is a representative algorithm for the straight line detection, a great number of edge pixels generated from noisy or complex images cause enormous amount of computation and pseudo straight lines. This paper proposes a two step straight line detection algorithm to improve the conventional Hough transform. In the first step, the proposed algorithm divides an image into non-overlapping blocks and detects the information related to the straight line of the edge pixels in the block using a principal component analysis (PCA). In the second step, it detects the straight lines by performing the Hough transform limited slope area to the pixels associated with the straight line. Simulation results show that the proposed algorithm reduces average of ${\rho}$ computation by 94.6% and prevents the pseudo straight lines although some additional computation is needed.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.29 no.6
• /
• pp.464-472
• /
• 2018

Future spaceborne satellite synthetic aperture radar(SAR) system is expected to acquire high-resolution images over wide swath areas. Conventional SAR systems suffer from ambiguity problems in both azimuth and range directions that lead to image quality degradation. Recently, multiple input multiple output(MIMO) SAR techniques having multiple orthogonal waveforms are proposed to overcome the conventional ambiguity problems in wide-swath imaging modes. In this paper, noisy orthogonal frequency division multiplex(OFDM) waveforms are developed to reduce the ambiguity problems and suppress the image quality degradation. SAR simulations are performed to evaluate the performance of the proposed technique for wide-swath SAR imaging.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.7
• /
• pp.2633-2648
• /
• 2015

The problem of visual words' synonymy and ambiguity always exist in the conventional bag of visual words (BoVW) model based object category methods. Besides, the noisy visual words, so-called "visual stop-words" will degrade the semantic resolution of visual dictionary. In view of this, a novel bag of visual words method based on PLSA and chi-square model for object category is proposed. Firstly, Probabilistic Latent Semantic Analysis (PLSA) is used to analyze the semantic co-occurrence probability of visual words, infer the latent semantic topics in images, and get the latent topic distributions induced by the words. Secondly, the KL divergence is adopt to measure the semantic distance between visual words, which can get semantically related homoionym. Then, adaptive soft-assignment strategy is combined to realize the soft mapping between SIFT features and some homoionym. Finally, the chi-square model is introduced to eliminate the "visual stop-words" and reconstruct the visual vocabulary histograms. Moreover, SVM (Support Vector Machine) is applied to accomplish object classification. Experimental results indicated that the synonymy and ambiguity problems of visual words can be overcome effectively. The distinguish ability of visual semantic resolution as well as the object classification performance are substantially boosted compared with the traditional methods.