Matching Stage Research Articles

TopicFM+: Boosting Accuracy and Efficiency of Topic-Assisted Feature Matching.

This study tackles image matching in difficult scenarios, such as scenes with significant variations or limited texture, with a strong emphasis on computational efficiency. Previous studies have attempted to address this challenge by encoding global scene contexts using Transformers. However, these approaches have high computational costs and may not capture sufficient high-level contextual information, such as spatial structures or semantic shapes. To overcome these limitations, we propose a novel image-matching method that leverages a topic-modeling strategy to capture high-level contexts in images. Our method represents each image as a multinomial distribution over topics, where each topic represents semantic structures. By incorporating these topics, we can effectively capture comprehensive context information and obtain discriminative and high-quality features. Notably, our coarse-level matching network enhances efficiency by employing attention layers only to fixed-sized topics and small-sized features. Finally, we design a dynamic feature refinement network for precise results at a finer matching stage. Through extensive experiments, we have demonstrated the superiority of our method in challenging scenarios. Specifically, our method ranks in the top 9% in the Image Matching Challenge 2023 without using ensemble techniques. Additionally, we achieve an approximately 50% reduction in computational costs compared to other Transformer-based methods. Code is available at https://github.com/TruongKhang/TopicFM.

A Real-Time Global Re-Localization Framework for a 3D LiDAR-Based Navigation System

Place recognition is widely used to re-localize robots in pre-built point cloud maps for navigation. However, current place recognition methods can only be used to recognize previously visited places. Moreover, these methods are limited by the requirement of using the same types of sensors in the re-localization process and the process is time consuming. In this paper, a template-matching-based global re-localization framework is proposed to address these challenges. The proposed framework includes an offline building stage and an online matching stage. In the offline stage, virtual LiDAR scans are densely resampled in the map and rotation-invariant descriptors can be extracted as templates. These templates are hierarchically clustered to build a template library. The map used to collect virtual LiDAR scans can be built either by the robot itself previously, or by other heterogeneous sensors. So, an important feature of the proposed framework is that it can be used in environments that have never been visited by the robot before. In the online stage, a cascade coarse-to-fine template matching method is proposed for efficient matching, considering both computational efficiency and accuracy. In the simulation with 100 K templates, the proposed framework achieves a 99% success rate and around 11 Hz matching speed when the re-localization error threshold is 1.0 m. In the validation on The Newer College Dataset with 40 K templates, it achieves a 94.67% success rate and around 7 Hz matching speed when the re-localization error threshold is 1.0 m. All the results show that the proposed framework has high accuracy, excellent efficiency, and the capability to achieve global re-localization in heterogeneous maps.

OS-PSO: A Modified Ratio of Exponentially Weighted Averages-Based Optical and SAR Image Registration.

Optical and synthetic aperture radar (SAR) images exhibit non-negligible intensity differences due to their unique imaging mechanisms, which makes it difficult for classical SIFT-based algorithms to obtain sufficiently correct correspondences when processing the registration of these two types of images. To tackle this problem, an accurate optical and SAR image registration algorithm based on the SIFT algorithm (OS-PSO) is proposed. First, a modified ratio of exponentially weighted averages (MROEWA) operator is introduced to resolve the sudden dark patches in SAR images, thus generating more consistent gradients between optical and SAR images. Next, we innovatively construct the Harris scale space to replace the traditional difference in the Gaussian (DoG) scale space, identify repeatable key-points by searching for local maxima, and perform localization refinement on the identified key-points to improve their accuracy. Immediately after that, the gradient location orientation histogram (GLOH) method is adopted to construct the feature descriptors. Finally, we propose an enhanced matching method. The transformed relation is obtained in the initial matching stage using the nearest neighbor distance ratio (NNDR) and fast sample consensus (FSC) methods. And the re-matching takes into account the location, scale, and main direction of key-points to increase the number of correctly corresponding points. The proposed OS-PSO algorithm has been implemented on the Gaofen and Sentinel series with excellent results. The superior performance of the designed registration system can also be applied in complex scenarios, including urban, suburban, river, farmland, and lake areas, with more efficiency and accuracy than the state-of-the-art methods based on the WHU-OPT-SAR dataset and the BISTU-OPT-SAR dataset.

Uncovering merchants’ willingness to wait in on-demand food delivery markets

While traditional on-demand food delivery services help restaurants reach more customers and enable doorstep deliveries, they also come with drawbacks, such as high commission fees and limited control over the delivery process. White-label food delivery services have emerged as an alternative, ready-to-use platform for restaurants to arrange delivery for customer orders received through their applications or websites, without the constraints imposed by traditional on-demand food delivery platforms or the need to develop an in-house delivery operation. Although several studies have investigated consumer behavior when using traditional on-demand food delivery services, there is limited research on merchants’ behavior when adopting white-label food delivery services. In this research, we develop a non-parametric survival model to estimate merchants’ willingness to wait when using white-label food delivery services and examine how various factors, such as delivery fees, the number of placed orders, and average waiting time, affect merchants’ willingness to wait, drawing on a dataset of both delivered and canceled orders from a crowd-sourcing delivery platform in Singapore. The empirical results show that merchants’ willingness to wait has a non-linear relationship with their average waiting time; it initially increases and then decreases with average waiting time. Moreover, the relationship between merchants’ willingness to wait during the pick-up stage and their average waiting times in the matching stage follows a similar non-linear trend. Merchants who have experienced lengthy waiting times on average in the matching stage tend to be less patient in the pick-up stage. This research sheds light on the stage-specific dynamics of merchants waiting behavior in white-label delivery service and provides insights for delivery platforms to optimize their operational strategies and enhance user experiences.

Evaluating the Effectiveness of Penalty Kicks in the UEFA Champions League: Optimal Techniques and Directions for Scoring Success

This study examined the strategic execution of penalty kicks in professional soccer, focusing on the high-stakes environment of the UEFA Champions League. Utilizing a dataset of 243 penalty kicks from five seasons (2018/2019–2022/2023), sourced from YouTube and cross-referenced with Transfermarkt, we employed a detailed nine-zone goal classification system to analyze shot distribution and success rates. We investigated various factors influencing penalty outcomes, including match stage, home-field advantage, minute of the match, current score, player status, previous score, footedness, and rebound goals. Findings indicated that shots aimed at the center and upper areas of the goal have the highest success rates. It was also found that home advantage considerably enhances scoring chances in both group and knockout stages. This research fills a gap in the literature by providing insights into penalty kick strategies in the UEFA Champions League, offering valuable implications for coaching practices, player training, and analytical frameworks in high-pressure scenarios.

Video text tracking with transformer-based local search

Video text tracking is a highly significant branch in the field of multi-object tracking (MOT), aiming to detect all text instances in video frames and construct trajectories for each text. Conventional text tracking methods typically follow the Tracking-By-Detection (TbD) paradigm which involves two separate steps of detection and association. In complex scenarios, abundant detection omission of text instances is caused by exposure, occlusion, motion blur, and the like, leading to text trajectory breakpoints in the TbD methods. To this end, video text tracking with transformer-based local search (LSTrack) is proposed in this paper. We utilize historical trajectory information to estimate the approximate search areas where omitted texts are located. In this case, our local search tracker leverages the text images in historical trajectories as references to directly recall text instances in the search areas. In addition, our tracking framework can combine the explicit semantic information output from an OCR model to obtain the semantic version (SV) of LSTrack, which for the first time uses the text edit distance as the distance measurement in matching stages to achieve better text tracking results. Ultimately, our method achieves an advanced performance on several public benchmarks. In particular, LSTrack (SV) achieves state-of-the-art performance on the Minetto benchmark.

TIAM COFFEE SHOP MARKETING STRATEGY USING THE SWOT ANALYSIS METHOD

The aim of this research is to find out the SWOT matrix (IFAS and EFAS) at Kedai Kopi Tiam and find out alternative marketing strategies. This research uses the SWOT analysis method, namely strategic planning which is used to evaluate strengths, weaknesses, opportunities and threats. The SWOT model used in this research is a strategy for formulating an analytical framework which is divided into 3 stages, namely the input stage, matching stage and decision stage. Based on the results of the data analysis obtained, it was concluded that Kedai Kopi Tiam occupies quadrant I, which means the company is growing and developing. In this strategy, of course Kedai Kopi Tiam must maintain its strengths and take advantage of existing opportunities to avoid external threats and weaknesses within the company.

Temporal subgraph matching method for multi-connected temporal graph

Temporal graph matching is used to find a matching subgraph set that satisfies temporal and topological information on the temporal graph. It is widely used in real scenarios and provides important support for big data mining and analysis. Most existing matching methods take the single-connected graph as the matching graph, which has a high computational cost. Simultaneously, the construction and matching complexity of its index on multi-connected temporal subgraph matching is high. To solve the above problems, we propose a new temporal subgraph matching problem for multi-connected temporal graphs and proposes the corresponding efficient filtering-selecting-matching three-stage multi-connected temporal subgraph matching method. In the filtering stage, a vertex-based filter is proposed to generate a candidate temporal subgraph set. In the select phase, a set of temporal subgraphs satisfying the vertex mapping is found. In the matching stage, the single-filter temporal subgraph matching method and temporal subgraph matching based on the temporal information tree index method are proposed to solve the problems that the edges of multi-connected temporal subgraphs cannot be mapped and the single graph matching efficiency is low, respectively. The experimental results show that the proposed method improves the matching accuracy, and the corresponding index size is reduced.

A multi-task evolutionary algorithm for solving the problem of transfer targets

In recent years, multi-task optimisation, aimed at handling multiple optimisation problems simultaneously, has received great attention in the field of evolutionary algorithms. Research on multi-task evolutionary algorithms mostly focuses on solving the problem of effective transfer, which usually transfers the extracted effective knowledge to the matching selection stage or environment selection stage. However, the impact of these transfer targets on algorithm performance is rarely studied. To solve the problem of transfer targets, we propose a multi-task evolutionary algorithm with a hybrid knowledge transfer strategy (MTEA-HKTS). Firstly, we determined three transfer targets and designed knowledge transfer strategies for each stage by analysing the population changes in the differential evolution algorithm. Secondly, a knowledge transfer strategy is devised to control the transfer targets. Finally, we designed a new archive update strategy that uses a distribution model constructed for recent generations of populations to calculate gene similarity. The experimental results on the multi-task benchmark problems show that the transfer targets affect the performance of the multi-task optimisation algorithm, and verify the superiority of the proposed MTEA-HKTS algorithm.

Formulation of Risk Management Strategy for Sharia Mortgage Products during the Covid-19 Pandemic

The COVID-19 pandemic has had a significant impact on the development of the world economy, including the banking and financial services industry. One of the financing products at Bank BTN Syariah is the Home Ownership Financing-Mortgage (Sharia KPR), where the more financing funds are channeled, of course, the greater the potential for risk. The purpose of this study was to formulate a risk management strategy for Sharia KPR products that Bank BTN Syariah KC Bekasi can use during the Covid-19 pandemic. To analyze and develop the bank strategy, we use the Grand Strategy Matrix method by blending internal and external factors, where the analysis and strategy selection process is divided into three stages: Input stage, matching stage, and Decision stage. The results of the study showed that several internal and external factors occur in Bank BTN Sharia KC Bekasi that affect the development of Sharia KPR products; Bank BTN applies risk management based on the 5C principle to minimize the occurrence of risks that may occur. Managerial implications of this research are credit risk management, management, increased risk monitoring, effective communication, product innovation, conducting education and training, and Sharia compliance.

Predicting the Winner of a Twenty20 International Cricket Match: Classification and Explainable Machine Learning Approach

We present a supervised machine learning approach to predict the winner of a Twenty20 (T20) international match. The prediction dynamically changes as the match progresses. We also use explainable machine learning techniques (SHAP scores) to understand the importance of various features in making the decision at various stages of the T20 match. We present results on a dataset of 808 men's T20 international matches. The dynamic accuracy increases from about 55% in the initial stages of the T20 match to a maximum of about 85% in the final stages of the match (with an overall accuracy of about 63% in innings 1 and 74% in innings 2). SHAP scores reveal that team strength is an important feature in making the prediction in initial stages of the match; however, in the final stages, match situation plays the dominant role in the decision making process. Our work may help team coaches and captains to assess their chances of winning and/or chart a course towards winning in the ongoing T20 match, as well as be useful for sports analytics and gambling websites and apps.

Designing a Marketing Strategy for Bawis Habati Fish Chips Using the Quantitative Strategic Planning Matrix (QSPM) Method

Habati fish chips is one of the Micro, Small, and Medium Enterprises (MSME) in the food industry in the city of Bontang. Habati MSME still applies simple promotion methods and does not target the teenage market to compete. This research aims to design a marketing strategy for Habati fish chips in Bontang city, targeting a new market segment, which is teenagers. The marketing strategy analysis is conducted in three stages: input stage, matching stage, and decision stage. In the input stage, Internal Factor Evaluation (IFE) matrix is used with a score of 2.338, External Factor Evaluation (EFE) matrix with a score of 2.764, and Competitive Profile Matrix (CPM) to determine the position of Habati MSME compared to Pak Ucil and Abadi Rasa based on the perspective of potential consumers. The matching stage is then used to formulate marketing strategies using the Internal-External (IE) matrix and Strengths, Weaknesses, Opportunities, and Threats (SWOT) matrix. Based on the overall methods used, the final stage is the decision stage. The decision stage utilizes the Quantitative Strategic Planning Matrix (QSPM) to select prioritized alternative strategies that can be implemented by Habati MSME. The research results provide seven alternative strategies with one prioritized strategy having the highest Total Attractive Score (TAS), which is to increase product sales through online marketing, utilizing marketplaces, and promoting product advantages, with a TAS of 7.074.

Grounding rod hanging and removing robot with hand-eye self-calibration capability in substation

In this paper, a robot system for hanging and removing grounding rods is designed by integrating 3D recognition technology, hand-eye self-calibration technology, and automatic operation technology. Specifically, a novel hand-eye self-calibration algorithm is proposed that only uses common objects in the actual scene, which differs from traditional robot hand-eye calibration in that it requires a dedicated calibration board to assist with offline completion. Addressing the problem that existing self-calibration methods cannot be optimized as a whole, resulting in low accuracy and instability of the solution, a multi-stage objective function optimization self-calibration algorithm is proposed. An optimization method based on the minimization of re-projection error is designed to compensate for the results, which uses an efficient Oriented fast and rotated brief (ORB) feature extraction algorithm and introduces a scoring mechanism to retain more correct matching points in the feature matching stage. Experimental verifications are conducted using both public dataset and practical robot system. In the dataset experiment, our method demonstrates superior accuracy and robustness compared to existing self-calibration methods. Furthermore, the practical robot platform experiment confirms the feasibility and efficacy of our approach across various wind speeds and lighting conditions.

Poor cardiac output reserve in pulmonary arterial hypertension is associated with right ventricular stiffness and impaired interventricular dependence.

Pulmonary arterial hypertension (PAH) is characterised by poor exercise tolerance. The contribution of right ventricular (RV) diastolic function to the augmentation of cardiac output during exercise is not known. This study leverages pressure-volume (P-V) loop analysis to characterise the impact of RV diastology on poor flow augmentation during exercise in PAH. RV P-V loops were measured in 41 PAH patients at rest and during supine bike exercise. Patients were stratified by median change in cardiac index (CI) during exercise into two groups: high and low CI reserve. Indices of diastolic function (end-diastolic elastance (E ed)) and ventricular interdependence (left ventricular transmural pressure (LVTMP)) were compared at matched exercise stages. Compared to patients with high CI reserve, those with low reserve exhibited lower exercise stroke volume (36 versus 49 mL·m-2; p=0.0001), with higher associated exercise afterload (effective arterial elastance (E a) 1.76 versus 0.90 mmHg·mL-1; p<0.0001), RV stiffness (E ed 0.68 versus 0.26 mmHg·mL-1; p=0.003) and right-sided pressures (right atrial pressure 14 versus 8 mmHg; p=0.002). Higher right-sided pressures led to significantly lower LV filling among the low CI reserve subjects (LVTMP -4.6 versus 3.2 mmHg; p=0.0001). Interestingly, low exercise flow reserve correlated significantly with high afterload and RV stiffness, but not with RV contractility nor RV-PA coupling. Patients with poor exercise CI reserve exhibit poor exercise RV afterload, stiffness and right-sided filling pressures that depress LV filling and stroke work. High afterload and RV stiffness were the best correlates to low flow reserve in PAH. Exercise unmasked significant pathophysiological PAH differences unapparent at rest.

Coffee Shop Business Strategy Analysis Using the Business Model Canvas Method and Quantitative Strategic Planning Matrix (Case Study: Coffee Shop 'Kopi Dul')

Coffee Shop Kopi Dul is a Coffee Shop that has been established since 2018. The development of Coffee Shop which is getting higher from year to year requires Coffee Shop Kopi Dul to continue to innovate in order to survive amidst competition by mapping business models and creating business development strategies based on internal factors and external. The aim of this research is to plan the development of the Kopi Dul Coffee Shop business using the Business Model Canvas and QSPM methods. The data processing process in this research begins by creating a Business Model Canvas, the input stage is the IFE and EFE matrices, the matching stage is the IE, SPACE and SWOT matrices, the decision stage uses the QSPM method, finally a new Business Model Canvas which has been added to alternative strategies. obtained. The results of the research analysis show that the Kopi Dul Coffee Shop is in a hold and maintain position based on the results of the IE matrix, is in the aggressive quadrant based on the results of the SPACE matrix, and has 8 alternative strategy formulations based on the SWOT matrix. At the Decision stage, the alternative priority strategy is "Conducting regular competitor observations" with an STA value of 6.371. Next, a proposed Business Model Canvas is created by adding alternative strategy formulations at the matching stage.

Development Strategy Of Agro-tourism For Sustainability Of Agricultural Business Entity: Case Study Of Sawangan Farm In Depok, West Java

Abstract The lack of net income that farmers receive often puts pressure on the sustainability of agribusiness businesses. This research aims to analyze the agricultural business development strategies through business diversification at an agricultural and livestock business in Depok City called Sawangan Farm Dairy. This study analyses the factors of internal and external environmental conditions that affect the increase in income and plans alternative business development strategies. The research method is descriptive-qualitative using internal factor evaluation (IFE), external factor evaluation (EFE). Analysis method were SWOT analysis and Quantitative Strategies Planning Matrix (QSPM). Sampling technic used were purposive sampling. The results of the IFE analysis show a score of 3.00, which means it is in the high category. This means that the Sawangan Farm business, from an internal aspect, has strategic value to continue to develop. The results of the EFE analysis got a score of 3.32, which is also included in the high category. This can indicate that the company has a strong external position to take advantage of opportunities and overcome existing threats to achieve company goals. The simulation results of the IFE and EFE analysis scores, or the IE matching stage, show that the company is in quadrant I. For the SWOT analysis, six strategies were found, and from the QSPM analysis, the first rank to be implemented was conducting tourism marketing through online media to potential market areas, namely the cities nearest major cities, with a value of 6.88.

Prognostic Significance of Synchronous Colorectal Adenocarcinoma: A Matched-Pair Analysis.

To determine the prognostic significance of the synchronous colorectal cancer (S-CRC) on survival and recurrence rate. Authors conducted an analysis of 90 colorectal adenocarcinoma patients who received a curative (R0) resection with a full course of standard adjuvant treatment. A total of 45 patients diagnosed with S-CRC at the time of initial presentation were individually matched to a group of 45 solitary CRC patients in pair at a ratio of 1:1. The case-matched criteria included age (± 5 years), gender, tumor location, and tumor stage. For S-CRC, the most advanced pathologic lesion was defined as the index lesion, and the matching cancer stage was categorized according to the index lesion. The N-stage was determined based on all lymph nodes. There were a higher number of retrieved nodes in patients with S-CRC than those with solitary CRC. The median (min, max) of the total number of retrieved nodes for S-CRC was 18 (3, 53) nodes, compared to 14 (4, 45) nodes for solitary CRC (p < 0.01). All patients were without distant metastasis (stage I to III). The total accumulative number of patients experiencing tumor recurrence was 9 (20%) amongst the solitary CRC patients and 18 (40%) amongst the S-CRC patients at the 15-year surveillance period (p<0.05). The disease-free survival (DFS) (mean + SD) was 147.6 + 9.3 months in the solitary CRC group, compared to 110.5 + 11.7 months in the S-CRC group (p<0.05). Amongst S-CRC patients, those having primary and synchronous tumors located across anatomical segments had poorer DFS (70.5 months) and higher 15-year tumor recurrence rate (17.8%) than those with all tumors in the same or contiguous anatomical segments. In addition, the S-CRC patients with all tumors located in contiguous segment had a longer DFS (123.7 months) than the other types of anatomical correlation. Patients with S-CRC had worse prognosis than those with solitary CRC. For S-CRC, the anatomical correlation between the primary and the synchronous tumors may influence DFS and recurrence rate.

Strategy formulation to support automation and digitalization initiative: A case study of an aerospace component manufacturer in Indonesia

Purpose: Strategy formulation is part of organization design study to evaluate decision thoroughly, then enable organization to achieve company goal. This study aims to formulate strategy of automation and digitalization for PT XYZ, to be ready for welcoming future growth of demand by maintain competitive cost within the same industry. Method: The study collects primary data from focus group discussion with selected participant of PT XYZ, located in Bandung, Indonesia, with participant selection based on information rich person reference. Data collection and analysis utilize qualitative and quantitative approach, and utilize three steps of strategy formulation (input stage, matching stage and decision stage). Results: Strategy formulation steps as per theory, applicable to finalize strategy to support industry 4.0 transformation at PT XYZ. Focus Group Discussion able to collect organization aspiration, to ensure strategy will align to company objective. Combine steps with latest internal and external factor analysis method, such as PESTLE (Political Economy Social Technological Environment), VRIO (Value Rarity InImitable Organize), Porter Value Chain analysis and QSPM (Quantitative Strategic Planning Matrix) help broaden the perspective of the participant during discussion. The study resulted eight strategies, which then simplified by data reduction approach to be four. A decision of sequence implementation also agreed during focus group discussion with participants. Contributions: By conducting this study, author expect it will prove the practical implementation of strategy formulation in industry 4.0 transformation for aerospace industry in Indonesia and Asia.

Point-Sim: A Lightweight Network for 3D Point Cloud Classification

Analyzing point clouds with neural networks is a current research hotspot. In order to analyze the 3D geometric features of point clouds, most neural networks improve the network performance by adding local geometric operators and trainable parameters. However, deep learning usually requires a large amount of computational resources for training and inference, which poses challenges to hardware devices and energy consumption. Therefore, some researches have started to try to use a nonparametric approach to extract features. Point-NN combines nonparametric modules to build a nonparametric network for 3D point cloud analysis, and the nonparametric components include operations such as trigonometric embedding, farthest point sampling (FPS), k-nearest neighbor (k-NN), and pooling. However, Point-NN has some blindness in feature embedding using the trigonometric function during feature extraction. To eliminate this blindness as much as possible, we utilize a nonparametric energy function-based attention mechanism (ResSimAM). The embedded features are enhanced by calculating the energy of the features by the energy function, and then the ResSimAM is used to enhance the weights of the embedded features by the energy to enhance the features without adding any parameters to the original network; Point-NN needs to compute the similarity between each feature at the naive feature similarity matching stage; however, the magnitude difference of the features in vector space during the feature extraction stage may affect the final matching result. We use the Squash operation to squeeze the features. This nonlinear operation can make the features squeeze to a certain range without changing the original direction in the vector space, thus eliminating the effect of feature magnitude, and we can ultimately better complete the naive feature matching in the vector space. We inserted these modules into the network and build a nonparametric network, Point-Sim, which performs well in 3D classification tasks. Based on this, we extend the lightweight neural network Point-SimP by adding some trainable parameters for the point cloud classification task, which requires only 0.8 M parameters for high performance analysis. Experimental results demonstrate the effectiveness of our proposed algorithm in the point cloud shape classification task. The corresponding results on ModelNet40 and ScanObjectNN are 83.9% and 66.3% for 0 M parameters—without any training—and 93.3% and 86.6% for 0.8 M parameters. The Point-SimP reaches a test speed of 962 samples per second on the ModelNet40 dataset. The experimental results show that our proposed method effectively improves the performance on point cloud classification networks.

Gradient Weakly Sensitive Multi-Source Sensor Image Registration Method

Aiming at the nonlinear radiometric differences between multi-source sensor images and coherent spot noise and other factors that lead to alignment difficulties, the registration method of gradient weakly sensitive multi-source sensor images is proposed, which does not need to extract the image gradient in the whole process and has rotational invariance. In the feature point detection stage, the maximum moment map is obtained by using the phase consistency transform to replace the gradient edge map for chunked Harris feature point detection, thus increasing the number of repeated feature points in the heterogeneous image. To have rotational invariance of the subsequent descriptors, a method to determine the main phase angle is proposed. The phase angle of the region near the feature point is counted, and the parabolic interpolation method is used to estimate the more accurate main phase angle under the determined interval. In the feature description stage, the Log-Gabor convolution sequence is used to construct the index map with the maximum phase amplitude, the heterogeneous image is converted to an isomorphic image, and the isomorphic image of the region around the feature point is rotated by using the main phase angle, which is in turn used to construct the feature vector with the feature point as the center by the quadratic interpolation method. In the feature matching stage, feature matching is performed by using the sum of squares of Euclidean distances as a similarity metric. Finally, after qualitative and quantitative experiments of six groups of five pairs of different multi-source sensor image alignment correct matching rates, root mean square errors, and the number of correctly matched points statistics, this algorithm is verified to have the advantage of robust accuracy compared with the current algorithms.