Multilevel Matching Research Articles

Overlap-Aware Hierarchical Decoder for point cloud registration

Extracting high-quality correspondences is a critical challenge in current feature-learning based point cloud registration methods. Recently, coarse-to-fine network structures have shown great potential in addressing this challenge. Inspired by such structures, we investigate the effectiveness of two-stage network optimization in matching and propose a non-keypoint registration model called Overlap-Aware Hierarchical Decoder (OAH-Net). The construction of this model focuses on reducing outliers in the matching results and improving the understanding of geometric transformation invariance. To achieve this goal, we propose a point-to-point perception module for encoding the paired point clouds and perceiving the overlapping regions, as well as a pyramid hierarchical decoder for decoding multi-level features. We design an optimal matching mechanism adapted to the pyramid structure to handle accurate correspondences from multi-level matching, thereby improving the accuracy of the final registration. Experimental evaluations conducted on indoor and outdoor benchmark tests validate the outstanding performance of OAH-Net, which not only ensures low parameter count and computational efficiency but also demonstrates accuracy and stability comparable to state-of-the-art models.

A Spatial Clustering and Matching Game–Based Multihop Routing Algorithm for Heterogeneous UAVs

Unmanned aerial vehicle (UAV) clusters are increasingly deployed in military and civilian applications, necessitating efficient and reliable communication networks for cooperative operations. However, heterogeneous UAVs pose significant challenges for data transmission within and outside the cluster due to their high mobility, rapid topology changes, and limited node energy. For the routing planning problem of transmitting data from autonomous, heterogeneous UAVs to the cloud, existing methods struggle to accommodate the heterogeneity of UAVs and the communication bandwidth requirements, given the NP‐hard nature of the problem. To address this issue, this paper first constructs an exact model based on mixed‐integer linear programming to describe the heterogeneous UAVs, comprehensively searches the solution space, and generates a reasonable routing scheme for the heterogeneous UAVs, guiding the cluster communication network system design. Secondly, a heuristic algorithm is proposed that leverages density–based spatial clustering and matching game theory. The algorithm calculates relay nodes based on local density and relative distance. It generates an approximate optimal data transmission path for each UAV through a multilevel matching process, effectively reducing communication delay within the cluster. Finally, the paper conducts simulation experiments in randomly generated regional scenarios to validate the efficiency and effectiveness of the proposed method.

Community exposure to armed conflict and subsequent onset of alcohol use disorder.

To measure the independent consequences of community-level armed conflict beatings on alcohol use disorders (AUD) among males in Nepal during and after the 2000-2006 conflict. A population-representative panel study from Nepal, with precise measures of community-level violent events and subsequent individual-level AUD in males. Females were not included because of low AUD prevalence. Chitwan, Nepal. Four thousand eight hundred seventy-six males from 151 neighborhoods, systematically selected and representative of Western Chitwan. All residents aged 15-59 were eligible (response rate 93%). Measures of beatings in the community during the conflict (2000-2006), including the date and distance away, were gathered through neighborhood reports, geo-location and official resources, then linked to respondents' life histories of AUD (collected in 2016-2018) using the Nepal-specific Composite International Diagnostic Interview with life history calendar. Beatings nearby predict the subsequent onset of AUD during and after the armed conflict. Data were analyzed in 2021-2022. Cohort-specific, discrete-time models revealed that within the youngest cohort (born 1992-2001), those living in neighborhoods where armed conflict beatings occurred were more likely to develop AUD compared with those in other neighborhoods (odds ratio = 1.66; 95% confidence interval = 1.02-2.71). In this cohort, a multilevel matching analysis designed to simulate a randomized trial showed the post-conflict incidence of AUD for those living in neighborhoods with any armed conflict beatings was 9.5% compared with 5.3% in the matched sample with no beatings. Among male children living in Chitwan, Nepal during the 2000-2006 armed conflict, living in a neighborhood where armed conflict beatings occurred is associated with increased odds of developing subsequent alcohol use disorder. This association was independent of personal exposure to beatings and other mental disorders.

Single Object Tracking in Satellite Videos Based on Feature Enhancement and Multi-Level Matching Strategy

Despite significant advancements in remote sensing object tracking (RSOT) in recent years, achieving accurate and continuous tracking of tiny-sized targets remains a challenging task due to similar object interference and other related issues. In this paper, from the perspective of feature enhancement and a better feature matching strategy, we present a tracker SiamTM specifically designed for RSOT, which is mainly based on a new target information enhancement (TIE) module and a multi-level matching strategy. First, we propose a TIE module to address the challenge of tiny object sizes in satellite videos. The proposed TIE module goes along two spatial directions to capture orientation and position-aware information, respectively, while capturing inter-channel information at the global 2D image level. The TIE module enables the network to extract discriminative features of the targets more effectively from satellite images. Furthermore, we introduce a multi-level matching (MM) module that is better suited for satellite video targets. The MM module firstly embeds the target feature map after ROI Align into each position of the search region feature map to obtain a preliminary response map. Subsequently, the preliminary response map and the template region feature map are subjected to the Depth-wise Cross Correlation operation to get a more refined response map. Through this coarse-to-fine approach, the tracker obtains a response map with a more accurate position, which lays a good foundation for the prediction operation of the subsequent sub-networks. We conducted extensive experiments on two large satellite video single-object tracking datasets: SatSOT and SV248S. Without bells and whistles, the proposed tracker SiamTM achieved competitive results on both datasets while running at real-time speed.

AMC: Adaptive Multi-expert Collaborative Network for Text-guided Image Retrieval

Text-guided image retrieval integrates reference image and text feedback as a multimodal query to search the image corresponding to user intention. Recent approaches employ multi-level matching, multiple accesses, or multiple subnetworks for better performance regardless of the heavy burden of storage and computation in the deployment. Additionally, these models not only rely on expert knowledge to handcraft image-text composing modules but also do inference by the static computational graph. It limits the representation capability and generalization ability of networks in the face of challenges from complex and varied combinations of reference image and text feedback. To break the shackles of the static network concept, we introduce the dynamic router mechanism to achieve data-dependent expert activation and flexible collaboration of multiple experts to explore more implicit multimodal fusion patterns. Specifically, we construct AMC, our A daptive M ulti-expert C ollaborative network, by using the proposed router to activate the different experts with different levels of image-text interaction. Since routers can dynamically adjust the activation of experts for the current samples, AMC can achieve the adaptive fusion mode for the different reference image and text combinations and generate dynamic computational graphs according to varied multimodal queries. Extensive experiments on two benchmark datasets demonstrate that due to benefits from the image-text composing representation produced by an adaptive multi-expert collaboration mechanism, AMC has better retrieval performance and zero-shot generalization ability than the state-of-the-art method while keeping the lightweight model and fast retrieval speed. Moreover, we analyze the visualization of path activation, attention map, and retrieval results to further understand the routing decisions and semantic localization ability of AMC. The codes and pretrained models are available at https://github.com/KevinLight831/AMC .

A Software Tutorial for Matching in Clustered Observational Studies

Abstract: Many interventions occur in settings where treatments are applied to groups. For example, a math intervention may be implemented for all students in some schools and withheld from students in other schools. When such treatments are non-randomly allocated, researchers can use statistical adjustment to make treated and control groups similar in terms of observed characteristics. Recent work in statistics has developed a form of matching, known as multilevel matching, that is designed for contexts where treatments are clustered. In this article, we provide a tutorial on how to analyze clustered treatment using multilevel matching. We use a real data application to explain the full set of steps for the analysis of a clustered observational study.

Costs and Effects of School-Based Licensed Practical Nurses on Elementary Student Attendance and Chronic Absenteeism

Despite increasing reliance on licensed practical nurses (LPNs) to provide health services in schools, we do not know whether this is a cost-effective prevention strategy against student absenteeism. Therefore, we evaluated the costs and effectiveness of an LPN-based school nursing program for improving attendance and chronic absenteeism at a large, urban school district in the southeastern USA. We first identified a matched set of 46 elementary schools (23 nurse, 23 no-nurse) by using an optimal multilevel matching algorithm based on student- and school-level characteristics. We then conducted a cost-effectiveness analysis on the matched set, using the ingredients method to estimate societal costs and multilevel regression to estimate effects. The results indicated that despite substantial incremental costs of $68,228 per school, the presence of a full-time LPN was associated with at best negligible improvements, and at worst slight disimprovements, in attendance and chronic absenteeism. We recommend a careful review of the theory of change for LPN-based school nursing programs to clarify the specific inputs and activities that are expected to lead to improved student outcomes. Education agencies should develop explicit assignment, training, monitoring, and auditing plans to ensure LPNs are equitably distributed and that their activities are aligned with the theory of change. Education agencies should also explore whether expanded Medicaid billing can reduce their share of the nursing cost burden.

The Effect of Grandparenting on the Depression and Life Satisfaction among Middle-Aged and Older Chinese Adults.

Given the prevalence of depressive mental health symptoms among Chinese adults of grandparenting age in recent decades, a better understanding of how depression and life satisfaction among middle-aged and older adults in China are affected by their role as grandparents is called for. This study examines the relationship between grandparenting and depression and life satisfaction among Chinese adults using multilevel regression models based on a multilevel matching dataset formulated from the 2018 China Health and Retirement Longitudinal Study (CHARLS) and the 2018 China City Statistical Yearbook. The results show that for adults who take care of their grandchildren, living with their children can significantly reduce depression. Meanwhile, whereas spending more time taking care of grandchildren can lower life satisfaction, taking care of more grandchildren is related to higher life satisfaction. The findings of this study should help policymakers improve the quality of life of Chinese adults through better-targeted approaches.

Enhanced distance-aware self-attention and multi-level match for sentence semantic matching

Sentence semantic matching is a core research area in natural language processing, which is widely used in various natural language tasks. In recent years, attention mechanism has shown good performance in deep neural networks for sentence semantic matching. Most of the attention-based deep neural networks focus on sentences interaction which ignore modeling the core semantic of the sentence. In other words, they do not consider the importance of the relative distance of words when modeling the sentence semantics, which leads to deviations in modeling the core semantics of the sentence and unstable sentence interaction. Usually, people tend to associate words that are relatively close together when they read and believe that there is a deeper connection between them. Besides, the current interactive matching method after sentence modeling is relatively simple and it may be inadequate. In this paper, we build a well-performed distance-aware self-attention and multi-level matching model (DSSTM) for sentence semantic matching tasks. By considering the importance of different distance tokens, it can get the better original semantics of sentences and hold interactive matching method in multiple level after sentence modeling. To be specific, given two input sentences, we first encode them as contextual embeddings. Then, the contextual embeddings are handled by enhanced distance-aware self-attention to further strengthen the sentence semantic modeling from the whole and local aspect. At the same time, we apply the co-attention layer to extract cross-sentence interaction features while simplifying all the remaining components. Finally, we fuse them into the multi-level matching function to obtain the aggregation vector and learn divers matching representations, which is helpful to capture the diversity of sentence pairs. We conduct experiments on three sentence semantic matching tasks. Experimental results on these public datasets demonstrate that our model outperforms competitive baseline methods and our model has fewer parameters. Our source code is publicly availableat https://github.com/xiaodeng-1/DSSTM.

Application of Pattern Matching in Heating Furnace – Rough Rolling Process

Hot continuous rolling is a complex industrial process. Aiming at the problems of difficult optimization control, information island between process and strong coupling between parameters in 2250 mm rolling line of a steel plant, a parameter optimization method based on multi-level pattern matching is proposed, which is applied to the heating furnace – rough rolling link to adapt the best operation parameters in the current superior operational pattern library so as to improve the product quality of transfer bar. For the situation that the multi-level matching method cannot find the optimal operation mode, an input-output prediction model based on fuzzy neural network is established and particle swarm optimization algorithm is used until the output optimal operation parameters meet the requirements, and updates the new optimal mode to the superior operational pattern library to complete the expansion of the library. The simulation results show that the minimum relative error between the optimization results and the actual operation parameters is about 0.44%, which proves the feasibility of the pattern matching and evolution strategy, so as to achieve the purpose of improving the quality of hot rolling products and improving the enterprise efficiency.

Community-Level Social Support Infrastructure and Adult Onset of Major Depressive Disorder in a South Asian Postconflict Setting

Individual-level social support protects against major depressive disorder (MDD) among adults exposed to trauma. Little is known about the consequences of community-level interventions in the general population. To determine the potential consequences of neighborhood social infrastructure on incident MDD in a high-risk general population. This longitudinal, multilevel study estimated associations between a neighborhood-level program in a case-control design and subsequent individual outcomes across 10 years (2006-2015) in a cohort of young adults. Exogenously placed social programs simulate natural experiment conditions in a high-poverty population experiencing armed conflict (1998-2006). The western Chitwan valley in Nepal has a general population at high risk of MDD, with neighborhoods exposed to interventions to improve social support. From a random sample (response rate 93%) selected to represent the general population in 2016, participants aged 25 to 34 years in 2006 were studied. These individuals resided within 149 neighborhoods that varied in their availability of active social support programs. The analyses were conducted between October 2020 and November 2021. The Small Farmers Development Program (SFDP) uses shared, joint liability financial credit among neighbors to build social capital and cohesion within neighborhoods. Onset of DSM-IV MDD after the conflict, assessed by the Nepal-specific, clinically validated World Mental Health Composite International Diagnostic Interview with a life history calendar. The hypothesis tested was that exposure to SFDP reduced adult onset of MDD. Of the 1917 survey participants, 886 (46.2%) were women, and 856 (44.7%) were of Brahmin or Chhetri ethnicity. Of the 149 neighborhoods, 21 had an active SFDP group, and 156 of 1917 (8.1%) participants experienced MDD between 2006 and 2015. Discrete-time hazard models showed participants living in neighborhoods with an SFDP experienced incident MDD at nearly half the rate as others (odds ratio = 0.55; 95% CI, 0.30-1.02; P = .06). A multivariate, multilevel matching analysis showed the incidence of MDD among adults living in neighborhoods with an SFDP was 19 of 256 (7.4%), compared with 33 of 256 (12.9%) in the matched sample with no SFDP (z = 2.05; P = .04). Living in a neighborhood with community-level social support infrastructure was associated with reduced subsequent rates of adult-onset MDD, even in this high-risk population. Investments in such infrastructure may reduce population-level MDD, supporting clinical focus on potentially unpreventable cases.

Learning and Integrating Multi-Level Matching Features for Image-Text Retrieval

Learning and Integrating Multi-Level Matching Features for Image-Text Retrieval

Matching Methods for Clustered Observational Studies in Education

Many interventions in education occur in settings where treatments are applied to groups. For example, a reading intervention may be implemented for all students in some schools and withheld from students in other schools. When such treatments are nonrandomly allocated, outcomes across the treated and control groups may differ due to the treatment or due to baseline differences between groups. When this is the case, researchers can use statistical adjustment to make treated and control groups similar in terms of observed characteristics. Recent work in statistics has developed matching methods designed for contexts where treatments are clustered. This form of matching, known as multilevel matching, may be well suited to many education applications where treatments are assigned to schools. In this article, we provide an extensive evaluation of multilevel matching and compare it to multilevel regression modeling. We evaluate multilevel matching methods in two ways. First, we use these matching methods in a within-study comparison design and attempt to recover treatment effect estimates from three clustered randomized trials. Second, we conduct a simulation study. We find evidence that generally favors an analytic approach to statistical adjustment that combines multilevel matching with regression adjustment. We conclude with an empirical application.

Effect of ambient air quality on subjective well-being among Chinese working adults

In the context of rapid urbanization and climate change in China, it is especially important to understand the effects of air quality on the subjective well-being of working adults because of their prolonged exposure to air pollutants. This study examines the relationship between ambient air pollutant emission levels and the subjective well-being of working adults using multilevel regression models based on a multilevel matching dataset formulated from the 2016 China Labor-Force Dynamics Survey, the 2016 China City Statistical Yearbook, and the 3rd-Quarter 2016 air pollution index released by the China National Environmental Monitoring Centre. In particular, we examine whether these air pollutants affect subjective well-being and compare the results by age, population density, and regional settings in China. The results show that increased CO emissions compromise the happiness of working adults (β = −0.1574, p < 0.05 for happiness; β = −0.1759, p < 0.01 for life satisfaction). This negative effect is substantially ameliorated among older working adults (β = 0.0002), but is significantly stronger in medium-density (β = −0.2521 for happiness; β = −0.2451 for life satisfaction) and high-density cities (β = −0.3997 for happiness; β = −0.5736 for life satisfaction). In addition, CO emissions have less effect on life satisfaction for adult workers living in western China (β = −0.1051). The findings of this study have a strong potential to help policymakers improve the quality of life of residents through better-targeted approaches.

Learning Fine-Grained Fact-Article Correspondence in Legal Cases

Automatically recommending relevant law articles to a given legal case has attracted much attention as it can greatly release human labor from searching over the large database of laws. However, current researches only support coarse-grained recommendation where all relevant articles are predicted as a whole without explaining which specific fact each article is relevant with. Since one case can be formed of many supporting facts, traversing over them to verify the correctness of recommendation results can be time-consuming. We believe that learning fine-grained correspondence between each single fact and law articles is crucial for an accurate and trustworthy AI system. With this motivation, we perform a pioneering study and create a corpus with manually annotated fact-article correspondences. We treat the learning as a text matching task and propose a multi-level matching network to address it. To help the model better digest the content of law articles, we parse articles in form of premise-conclusion pairs with random forest. Experiments show that the parsed form yielded better performance and the resulting model surpassed other popular text matching baselines. Furthermore, we compare with previous researches and find that establishing the fine-grained fact-article correspondences can improve the recommendation accuracy by a large margin. Our best system reaches an F1 score of 96.3%, making it of great potential for practical use. It can also significantly boost the downstream task of legal decision prediction, increasing the F1 score by up to 12.7%. The dataset and code will be released upon acceptance. Code and dataset are available at https://github.com/gjdnju/MLMN.

Electromagnetic Inverse Scattering With Perceptual Generative Adversarial Networks

In this work, we introduce a learning-based method to achieve high-quality reconstructions for inverse scattering problems (ISPs). Particularly, the proposed method decouples the full-wave reconstruction model into two steps, including coarse imaging of dielectric profiles by the back-propagation scheme, and a resolution enhancement of coarse results as an image-to-image translation task solved by a novel perceptual generative adversarial network (PGAN). A perceptual adversarial (PA) loss, which is defined as a perceptual loss for the generator network using hidden layers from the discriminator network, is employed as a structural regularization in PGAN. The PA loss is further combined with the pixel-wise loss, and also possibly the adversarial loss, to enforce a multi-level match between the reconstructed image and its reference one. The adversarial training of the generator and discriminator networks ensures that the structural features of targets are dynamically learned by the generator. Numerical tests on both synthetic and experimental data verify that the proposed method is highly efficient and it achieves superior imaging results compared to other data-driven methods. The validation of the proposed PGAN on ISPs also provides a fast and high-precision way for solving other physics-related imaging problems.

A Novel Change Detection Framework in Urban Area Using Multilevel Matching Feature and Automatic Sample Extraction Strategy

Urban change detection (CD) using remote sensing images is of great significance for monitoring and analyzing the spatial and temporal distribution of changes within cities, which can be used to guide urban management. In order to reduce false detections and improve the CD automation, a novel CD framework for high resolution images is proposed in this article. First, a novel multilevel matching feature is presented, combining structural invariant features with multiscale dense matching features to comprehensively describe the invariant properties of ground objects at different levels. Second, a newly automatic training sample extraction strategy is proposed, in which sufficient and accurate no-change samples can be obtained by Gaussian-weighted Dempster–Shafer evidence theory and L1-norm, meanwhile, typical change samples can be extracted by sequential spectral change vector analysis. Utilizing the automatic extracted samples, the final CD results are obtained using four supervised classifiers, respectively ( k -nearest neighbor, support vector machine, rotation forest, and extra-trees). To validate the proposed CD framework, experiments are conducted in four datasets with spectral variability, spectral confusion between the changed objects and unchanged backgrounds, and misregistration. The results demonstrate that the proposed multilevel matching feature and automatic sample extraction strategy can obtain better results with different types of supervised classifiers, and can effectively improve automation of CD, which is applicable to the large spatial extent scene.

Clinical concept normalization with a hybrid natural language processing system combining multilevel matching and machine learning ranking.

Normalizing clinical mentions to concepts in standardized medical terminologies, in general, is challenging due to the complexity and variety of the terms in narrative medical records. In this article, we introduce our work on a clinical natural language processing (NLP) system to automatically normalize clinical mentions to concept unique identifier in the Unified Medical Language System. This work was part of the 2019 n2c2 (National NLP Clinical Challenges) Shared-Task and Workshop on Clinical Concept Normalization. We developed a hybrid clinical NLP system that combines a generic multilevel matching framework, customizable matching components, and machine learning ranking systems. We explored 2 machine leaning ranking systems based on either ensemble of various similarity features extracted from pretrained encoders or a Siamese attention network, targeting at efficient and fast semantic searching/ranking. Besides, we also evaluated the performance of a general-purpose clinical NLP system based on Unstructured Information Management Architecture. The systems were evaluated as part of the 2019 n2c2 challenge, and our original best system in the challenge obtained an accuracy of 0.8101, ranked fifth in the challenge. The improved system with newly designed machine learning ranking based on Siamese attention network improved the accuracy to 0.8209. We demonstrate the successful practice of combining multilevel matching and machine learning ranking for clinical concept normalization. Our results indicate the capability and interpretability of our proposed approach, as well as the limitation, suggesting the opportunities of achieving better performance by combining general clinical NLP systems.

Lifetime risk of primary shoulder arthroplasty over a decade: a population-level analysis

Purpose: Lifetime risk estimates can serve as a marker for disease burden in a population and enable health funders and the health workforce to plan for healthcare demand. While the lifetime risk of hip and knee arthroplasty have previously been estimated at national levels, this approach has not yet been applied to shoulder arthroplasty. This study aimed to quantify the lifetime risk of primary shoulder arthroplasty in Australia, and examine changes in lifetime risk over a ten-year period. Methods: De-identified individual-level data on all primary partial shoulder arthroplasty (PSA) and total shoulder arthroplasty (TSA) procedures performed in Australia from 1 January 2008 to 31 December 2017 were obtained from the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR). The AOANJRR has a near-complete capture rate (>98.8%) of all hip, knee and shoulder arthroplasties performed in Australia, and registry data are validated against state-level health department data using a sequential multi-level matching process. The full collection of AOANJRR national shoulder arthroplasty data commenced in 2008. Population data and life tables (incorporating life expectancy and all-cause mortality) were obtained from the Australian Bureau of Statistics. The lifetime risk of primary PSA and primary TSA was calculated for each year from 2008-2017 using AOANJRR data, population data and life tables data, according to a standardised formula. Separate calculations were undertaken for males and females. Results: From 2008 to 2017, a total of 38,868 primary shoulder arthroplasties were performed. Of these, 5,979 (15%) were PSA procedures and the majority (N=32,889, 85%) were TSA procedures. Glenohumeral osteoarthritis was the most common reason for both PSA (53% in 2017) and TSA (57% in 2017). The lifetime risk of PSA decreased significantly for both sexes over time. For males, lifetime risk decreased from 0.25% (95% CI 0.22% to 0.28%) in 2008 to 0.11% (95% CI 0.09% to 0.13%) in 2017. For females, lifetime risk decreased from 0.55% (95% CI 0.51% to 0.60%) to 0.11% (95% CI 0.09% to 0.13%). In contrast, there was a significant increase in the lifetime risk of TSA. For males, this tripled from 0.54% (95% CI 0.49% to 0.58%) in 2008 to 1.68% (95% CI 1.60% to 1.76%) in 2017. For females, lifetime risk more than doubled from 0.99% (95% CI 0.92% to 1.05%) to 2.77% (95% CI 2.67% to 2.88%). Conclusions: Compared to declining PSA trends, there was substantial growth in TSA use over a decade. To put these findings in context, the lifetime risk of TSA in Australia increased to 1 in 60 for males and 1 in 40 for females by the end of 2017. The steep increase in lifetime risk of TSA over time probably relates to addressing previously unmet need for surgery, given the increasing availability of skilled surgeons trained to perform the procedure and advances in prosthesis design and materials. Greater awareness among patients and health professionals of successful outcomes following TSA may have also contributed to the observed growth. These data improve our understanding of the rising national burden of primary TSA, and can assist in planning to meet future surgical demand. The methods used for this study can also be re-applied in future years for ongoing population-level surveillance and to facilitate a planned international comparison.

Efficient hybrid multi-level matching with diverse set of features for image retrieval

Content-based image retrieval has become popular in the retrieval of images from large image database using reduced human intervention. Researchers are still in need to develop effective systems for dealing many of the wide-scope scientific and medical applications. Past research works have faced a problem on differentiating different images by means of using the single features alone. In this paper, a multi-level matching scheme is introduced for retrieval of image based on a hybrid feature similarity integrating local and global features. Both global- and local-level features included in multi-level scheme are used for image representation. From an image, the color information is extracted globally using a new color, edge directivity descriptor and color-based features. Further, the interest of points from each image is detected using local descriptors called local binary pattern and speeded-up robust features. Using two image databases, the improved retrieval accuracy obtained with the combination of global and local features is analyzed. Experimental outcomes have revealed the effectiveness of proposed system on achieving 91% and 92% precision rates over two datasets compared to other existing methods.