• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.411-427
• /
• 2022

Digging historical and cultural information from seals in ancient books is of great significance. However, ancient Chinese seal samples are scarce and carving methods are diverse, and traditional digital image processing methods based on greyscale have difficulty achieving superior segmentation and recognition performance. Recently, some deep learning algorithms have been proposed to address this problem; however, current neural networks are difficult to train owing to the lack of datasets. To solve the afore-mentioned problems, we proposed an SEL-RefineMask which combines selector of feature pyramid network (SEL-FPN) with RefineMask to segment and recognize seals. We designed an SEL-FPN to intelligently select a specific layer which represents different scales in the FPN and reduces the number of anchor frames. We performed experiments on some instance segmentation networks as the baseline method, and the top-1 segmentation result of 64.93% is 5.73% higher than that of humans. The top-1 result of the SEL-RefineMask network reached 67.96% which surpassed the baseline results. After segmentation, a vision transformer was used to recognize the segmentation output, and the accuracy reached 91%. Furthermore, a dataset of seals in ancient Chinese books (SACB) for segmentation and small seal font (SSF) for recognition were established which are publicly available on the website.

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2384-2386
• /
• 2003

This paper presents an image processing method to recognize object's position from images for robot control applications. The Sobel Mask and the Dialated Gradient Mask algorithms are used to extract and refine the edge of the object. The inside of the extracted edge is filled with '1's for calculation of the center of mass to identify object's position. We propose a heuristic method which averages the positions of the objects in 10 frames to reject noises, and discuss the application of the proposed method to robot control.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.1795-1811
• /
• 2019

In order to achieve rapid and accurate detection of vehicle objects in complex traffic conditions, we propose a novel vehicle detection method. Firstly, more contextual and small-object vehicle information can be obtained by our Joint Feature Network (JFN). Secondly, our Evolved Region Proposal Network (EPRN) generates initial anchor boxes by adding an improved version of the region proposal network in this network, and at the same time filters out a large number of false vehicle boxes by soft-Non Maximum Suppression (NMS). Then, our Mask Network (MaskN) generates an example that includes the vehicle occlusion, the generator and discriminator can learn from each other in order to further improve the vehicle object detection capability. Finally, these candidate vehicle detection boxes are optimized to obtain the final vehicle detection boxes by the Fine-Tuning Network(FTN). Through the evaluation experiment on the DETRAC benchmark dataset, we find that in terms of mAP, our method exceeds Faster-RCNN by 11.15%, YOLO by 11.88%, and EB by 1.64%. Besides, our algorithm also has achieved top2 comaring with MS-CNN, YOLO-v3, RefineNet, RetinaNet, Faster-rcnn, DSSD and YOLO-v2 of vehicle category in KITTI dataset.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.7
• /
• pp.2547-2567
• /
• 2021

It has been widely acknowledged that occlusion impairments adversely distress many face recognition algorithms' performance. Therefore, it is crucial to solving the problem of face image occlusion in face recognition. To solve the image occlusion problem in face recognition, this paper aims to automatically de-occlude the human face majority or discriminative regions to improve face recognition performance. To achieve this, we decompose the generative process into two key stages and employ a separate generative adversarial network (GAN)-based network in both stages. The first stage generates an initial coarse face image without an occlusion mask. The second stage refines the result from the first stage by forcing it closer to real face images or ground truth. To increase the performance and minimize the artifacts in the generated result, a new refine loss (e.g., reconstruction loss, perceptual loss, and adversarial loss) is used to determine all differences between the generated de-occluded face image and ground truth. Furthermore, we build occluded face images and corresponding occlusion-free face images dataset. We trained our model on this new dataset and later tested it on real-world face images. The experiment results (qualitative and quantitative) and the comparative study confirm the robustness and effectiveness of the proposed work in removing challenging occlusion masks with various structures, sizes, shapes, types, and positions.

• Proceedings of the KSRS Conference
• /
• 1999.11a
• /
• pp.411-420
• /
• 1999

In geology, lineament features have been used to identify geological events, and many of scientists have been developed the algorithm that can be applied with the computer to recognize the lineaments. We choose several edge detection filter, line detection filters and Hough transform to detect an edge, line, and to vectorize the extracted lineament features, respectively. firstly the edge detection filter using a first-order derivative is applied to the original image In this step, rough lineament image is created Secondly, line detection filter is used to refine the previous image for further processing, where the wrong detected lines are, to some extents, excluded by using the variance of the pixel values that is composed of each line Thirdly, the thinning process is carried out to control the thickness of the line. At last, we use the Hough transform to convert the raster image to the vector one. A Landsat image is selected to extract lineament features. The result shows the lineament well regardless of directions. However, the degree of extraction of linear feature depends on the values of parameters and patterns of filters, therefore the development of new filter and the reduction of the number of parameter are required for the further study.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.711-714
• /
• 2024

In the dynamic landscape of modern machine learning, Federated Learning (FL) has emerged as a compelling paradigm designed to enhance privacy by enabling participants to collaboratively train models without sharing their private data. Specifically, Distillation-based Federated Learning, like Federated Learning with Model Distillation (FedMD), Federated Gradient Encryption and Model Sharing (FedGEMS), and Differentially Secure Federated Learning (DS-FL), has arisen as a novel approach aimed at addressing Non-IID data challenges by leveraging Federated Learning. These methods refine the standard FL framework by distilling insights from public dataset predictions, securing data transmissions through gradient encryption, and applying differential privacy to mask individual contributions. Despite these innovations, our survey identifies persistent vulnerabilities, particularly concerning the susceptibility to logit inversion attacks where malicious actors could reconstruct private data from shared public predictions. This exploration reveals that even advanced Distillation-based Federated Learning systems harbor significant privacy risks, challenging the prevailing assumptions about their security and underscoring the need for continued advancements in secure Federated Learning methodologies.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.4
• /
• pp.938-958
• /
• 2024

Unsupervised Video Object Segmentation (UVOS) is a highly challenging problem in computer vision as the annotation of the target object in the testing video is unknown at all. The main difficulty is to effectively handle the complicated and changeable motion state of the target object and the confusion of similar background objects in video sequence. In this paper, we propose a novel deep Dual-stream Co-enhanced Network (DC-Net) for UVOS via bidirectional motion cues refinement and multi-level feature aggregation, which can fully take advantage of motion cues and effectively integrate different level features to produce high-quality segmentation mask. DC-Net is a dual-stream architecture where the two streams are co-enhanced by each other. One is a motion stream with a Motion-cues Refine Module (MRM), which learns from bidirectional optical flow images and produces fine-grained and complete distinctive motion saliency map, and the other is an appearance stream with a Multi-level Feature Aggregation Module (MFAM) and a Context Attention Module (CAM) which are designed to integrate the different level features effectively. Specifically, the motion saliency map obtained by the motion stream is fused with each stage of the decoder in the appearance stream to improve the segmentation, and in turn the segmentation loss in the appearance stream feeds back into the motion stream to enhance the motion refinement. Experimental results on three datasets (Davis2016, VideoSD, SegTrack-v2) demonstrate that DC-Net has achieved comparable results with some state-of-the-art methods.

• Journal of Convergence for Information Technology
• /
• v.11 no.2
• /
• pp.31-40
• /
• 2021

Blockchain technology is also actively studied in the real estate transaction field, and real estate transactions have various ways. In this paper, we propose a model that simplifies the authentication procedure of auction systems using Ethereum's Ether to solve the problem of offline court auctions. The proposed model is written in Ethereum's Solidity language, the court registers the sale date and the sale date with the DApp browser, and the bidder accesses the address of the individual's wallet created through Metamask's private key. The bidder then selects the desired sale and enters the bid price amount to participate in the auction. The bidder's record of the highest bid price for the sale he wants is written on the Ethereum test network as a smart contract. and creates a block. Finally, smart contracts written on the network are distributed by the court auction manager to all nodes in the blockchain network, and each node in the blockchain network can be viewed and contract verified. As a result of analyzing the smart contracts of the proposed model and the performance of the system, there are fees incurred due to the creation and use of Ether on platforms using Ethereum, and participation. Ether's changes in value affect the price of the sale, resulting in inconsistent fees in smart contracts each time. However, in future work, we issue our own tokens to solve the market volatility problem and commission problem with the value change of Ether, and refine complex court auction systems.