Instance-level image retrieval aims to find images containing the same object as a given query, despite variations in size, position, or appearance. To address this challenging task, we propose Patchify, a simple yet effective patch-wise retrieval framework that offers high performance, scalability, and interpretability without requiring fine-tuning. Patchify divides each database image into a small number of structured patches and performs retrieval by comparing these local features with a global query descriptor, enabling accurate and spatially grounded matching. To assess not just retrieval accuracy but also spatial correctness, we introduce LocScore, a localization-aware metric that quantifies whether the retrieved region aligns with the target object. This makes LocScore a valuable diagnostic tool for understanding and improving retrieval behavior. We conduct extensive experiments across multiple benchmarks, backbones, and region selection strategies, showing that Patchify outperforms global methods and complements state-of-the-art reranking pipelines. Furthermore, we apply Product Quantization for efficient large-scale retrieval and highlight the importance of using informative features during compression, which significantly boosts performance.

Bitcoin proved the potential of blockchain technology through decentralized, transparent, and immutable transactions. However, there are still challenges for fast and stable transitions. Optimizing block propagation times within the network is one of them. Prolonged propagation times can restrict efficiency, scalability, and security. This paper presents a novel approach to reducing block propagation time through leveraging reinforce-ment learning (RL) for the node’s neighbor selection strategies. We implemented a Deep Q-Network (DQN) model in minimizing block receive times at each node, thereby impacting overall block propagation time. We used a model that defines node states based on latencies of outbound connections, which present the node’s region. By evaluating this model through simulations using SimBlock, a robust Bitcoin network simulator, we observed a significant reduction in block propagation time—approximately 30% for smaller networks and 20% for larger ones. Our analysis extended to node connections generated by our model and comparative evaluation against existing methodologies.

With the rise of NFTs, which serve as proof of ownership for assets, security tokens that enable transactions without intermediaries have become a popular topic. Tokenization eliminates the need for centralized markets and allows for fast trades. In addition to tokenization, there are attempts to increase asset liquidity by fractionalizing them, a concept known as fractional ownership. Despite the existence of some platforms that use tokenization and fractional ownership, there is still limited research on fractional NFTs, and institutional support is lacking. In this paper, we propose possible implementations of fractional NFTs and evaluate their gas costs, which are crucial for providing fractional NFT-related services. As most NFTs are minted based on the Ethereum blockchain, we implement fractional NFTs using ERC standards. Our evaluation shows that ERC-721 or ERC-1155 NFTs fractionalized into ERC-20 FTs have the lowest long-term gas costs.

In accordance with the guidelines of the Financial Action Task Force (FATF), Virtual Asset Service Providers (VASPs) should comply with a ‘travel rule’, which requires them to exchange originator’s and beneficiary’s personal information when transferring virtual assets. In this paper, we propose a novel blockchain-based travel rule compliance system that supports fully-decentralized data exchange. The proposed system uses a permissioned blockchain, and thereby eliminates the possibility of leakage of personal information to third parties or even to travel rule service providers, and ensures that travel rule data can be managed securely.

Since Bitcoin was launched, the blockchain technology and the cryptocurrencies have been in the spotlight. Ethereum, the second-generation blockchain introduced smart contracts, and many DApps have emerged due to them. Those DApps showed the feasibility of blockchain in various industries. However, even though the growth of blockchain technology, still many DApps are based on Ethereum, and supper its performance issue. Since the performance evaluation in Ethereum mainnet is almost impossible, and there are no formalized performance evaluation frameworks, it is hard to perform appropriate performance evaluation of Ethereum. Detail performance evaluations on Ethereum networks are essential for developing and operating DApps. In this paper, we use Hyperledger Caliper, an automated performance evaluation framework to evaluate an Ethereum private network, and the Ropsten testnet to overcome above problems. We evaluate the performance with a specific smart contract and analyze the results. Our evaluation results show that the Ethereum private network performs better than the Ropsten testnet, and the Ropsten testnet is unstable for performance evaluation. In addition, our results show that the performance of the transactions can differ following their content.

Photoacoustic imaging (PAI) has attracted great attention as a medical imaging method. Typically, photoacoustic (PA) images are reconstructed via beamforming, but many factors still hinder the beamforming techniques in reconstructing optimal images in terms of image resolution, imaging depth, or processing speed. Here, we demonstrate a novel deep learning PAI that uses multiple speed of sound (SoS) inputs. With this novel method, we achieved SoS aberration mitigation, streak artifact removal, and temporal resolution improvement all at once in structural and functional in vivo PA images of healthy human limbs and melanoma patients. The presented method produces high-contrast PA images in vivo with reduced distortion, even in adverse conditions where the medium is heterogeneous and/or the data sampling is sparse. Thus, we believe that this new method can achieve high image quality with fast data acquisition and can contribute to the advance of clinical PAI.

The prevalence of non-alcoholic fatty liver diseases (NAFLD) has increased steadily over the past decade. Thus, diagnosing NAFLD at the earliest stage, which is a reversible condition, has become increasingly important. Here, photothermal strain imaging (pTSI) is presented as a novel non-invasive tool for NAFLD diagnosis. The pTSI uses ultrasound to detect the difference in thermal strain between fat and water during a light-induced temperature rise, which is directly related to the pathological evidence of NAFLD. To demonstrate its feasibility, fat accumulation in in vivo rat livers is monitored non-invasively using pTSI, based on clinical ultrasound B-mode images. A total of 21 male Wistar rats of 3 weeks of age were prepared. Of these, 18 rats received methionine-choline deficient diet for 1 to 6 weeks (n = 3 per week) to induce NAFLD, whereas 3 rats received normal diet as controls (n = 3). Livers were heated by a lipid-sensitive continuous-wave laser, and strain was measured. Quantitative results from the pTSI were compared with histological analysis results using Oil-Red-O (ORO). The receiver operating characteristic curve of in vivo pTSI results for detecting moderate steatosis (ORO-stained area ≥33%) was constructed based on strain change rate measured in the liver region. The sensitivity and specificity of pTSI were 90% and 82%, respectively, and the area-under-the-curve was measured as 0.85 ± 0.03 (95% confidence interval). The pTSI results tested in the rodent NAFLD model showed great potential for pTSI to be used as a new diagnostic tool for NAFLD in the future.

We introduce a fast Nearest Neighbor Search (NNS) algorithm using an Approximate Cached k-d tree (ACk-d tree) structure for low dimensional data sets. The search process of the standard k-d tree starts from the root node and employs a tentative back-tracking search. In contrast, the proposed method begins to search at the appropriate leaf node (cached node) and applies a depth-first nontentative search. This method improves searching speed, with tradeoff of the searching accuracy. To get a proper starting node, the proposed method is based on two properties: i) The ithquery point is likely to be close to the (i-1)thquery point, ii) The ithquery point is likely to be close to the ithmodel point. These properties are rather right, in case of practical 3D point sets which are consecutively acquired from 3D point sensors (e.g. a stereo camera, the Kinect sensor, and LIDAR). Results show that the search time of the proposed method is superior to other variants of k-d tree for practical point data sets.

The Iterative Closest Point (ICP) algorithm is one of the most popular methods for geometric alignment of 3-dimensional data points. We focus on how to make it faster for 3D range scanner in intelligent vehicle. The ICP algorithm mainly consists of two parts: nearest neighbor search and estimation of transformation between two data sets. The former is the most time consuming process. Many variants of the k-d trees have been introduced to accelerate the search. This paper presents a remarkably efficient search procedure, exploiting two concepts of approximate nearest neighbor and local search. Consequently, the proposed algorithm is about 24 times faster than the standard k-d tree.

This paper presents a novel solution to overcome the disturbance noise (outlier) for the Extended Kalman Filter based Simultaneous Localization And Mapping (EKF-SLAM). The standard Kalman Filter (KF) is not robust to the disturbance noise. The possibility that disturbance may happen is high, because SLAM aims at exploring unknown environment. Hence KF based SLAM methods should consider how to handle the disturbance noise. Variations of KF have been introduced to overcome this problem. However, these methods employ manual parameter tuning, detecting/weighting method. The core of our algorithm is to inflate the state uncertainty by using the magnitude of innovation, without tuning and detecting. Although it is impossible to estimate the state value immediately, the inflated state uncertainty makes it possible for the estimated value to converge on the true value much faster. We evaluate the proposed method under the well-known benchmark Matlab program. The results show that the proposed method overcomes the disturbance noise and increases the performance of EKF-SLAM.

In this paper, we address a method for improving the accuracy of the feature map from the extended Kalman filter based SLAM (EKF SLAM) by estimating the systematic parameters of the robot. Most error of the robot while traveling is divided into two categories: systematic and non systematic error. The systematic error contributes much more to odometry errors than non systematic one on most smooth indoor surfaces. So, we appended the systematic parameters of the robot to the state vector of EKF SLAM as its elements and estimated the systematic parameters while performing the prediction and update state of EKF SLAM. Because the additional elements to be estimated are appended to the state vector of the EKF SLAM, this is called an augmented EKF SLAM (AEKF SLAM). Experimental result is presented to validate that our AEKF SLAM is able to generate a more accurate feature map than conventional EKF SLAM by decreasing odometric error of the robot.

This paper addresses the measurement noise of Extended Kalman Filter-based Simultaneous Localization And Mapping (EKF-SLAM). The Extended Kalman Filter (EKF) is based on the Gaussian noise with zero mean and should know the correct prior knowledge of control and measurement noise covariance matrices. If these conditions are not satisfied, EKF unavoidably diverges. The present paper proposes the method of a new adaptive kalman filter to be supported by Measurement Noise Estimator (MNE), which estimates the measurement noise distribution including biased noise and noise covariance, whenever the update step executes. We evaluate this method under well-known benchmark environment for SLAM problem. Simulation results show that the proposed algorithm overcomes degrading performance of the standard EKF under the condition of wrong knowledge of sensor statistics.