Filter-based Method Research Articles

Summary of PM2.5 Measurement Artifacts Associated with the Teledyne T640 PM Mass Analyzer Under Controlled Chamber Experimental Conditions Using Polydisperse Ammonium Sulfate Aerosols and Biomass Smoke

Particulate matter (PM) is a major primary pollutant emitted during wildland fires that has the potential to pose significant health risks to individuals/communities who live and work in areas impacted by smoke events. Limiting exposure is the principle measure available to mitigate health impacts of smoke and therefore the accurate determination of ambient PM concentrations during wildland fire events is critical to protecting public health. However, monitoring air pollutants in smoke impacted environments has proven challenging in that measurement interferences or sampling conditions can result in both positive and negative artifacts. The EPA has performed research on methods for the measurement of PM2.5 in a series of laboratory based studies including evaluation in smoke. This manuscript will summarize the results of the laboratory based evaluation of federal equivalent method (FEM) analyzers for PM2.5 with particular attention being given to the Teledyne-API Model T640 PM Mass analyzer, as compared to the filter-based federal reference method (FRM). The T640 is an optical-based PM analyzer and has been gaining wide use by state and local agencies in monitoring for PM2.5 U.S. National Ambient Air Quality Standards (NAAQS) attainment. At present, the T640 (includes both T640 and T640X) comprises ~40% of the PM2.5 FEM monitors in U.S. regulatory monitoring networks. In addition, the T640 has increasingly been employed for the higher time resolution comparison/evaluation of low-cost PM sensors including during smoke impacted events. Results from controlled non-smoke laboratory studies using generated ammonium sulfate aerosols, demonstrated a generally negative T640 measurement artifact that was significantly related to the PM2.5 concentration and particle size distribution. Results from biomass burning chamber studies demonstrated positive and negative artifacts significantly associated with PM2.5 concentration and optical wavelength dependent absorption properties of the smoke aerosol. Implications The results detailed in this product will provide state and local air monitoring agencies with the tools and knowledge to address PM2.5 measurement challenges in areas frequently impact by wildland fire smoke. The observed large positive and negative artifacts in the T640 PM mass determination has the potential to result in false exceedances of the PM2.5 NAAQS or in the disqualification of monitoring data through an exceptional event designation. In addition, the observed artifacts in smoke impacted air will have a detrimental effect on providing reliable public information when wildfires occur and also in identifying reference measurements for small sensor evaluation studies. Other PM2.5 FEMs such as the BAM-1022 perform better in smoke and are comparable to the filter based FRM. Care must be taken in choosing high time resolution FEM monitors that will be operated at smoke impacted sites. Accurate methods, such as the FRM and BAM-1022 will reduce the burden of developing and reviewing exceptional event request packages, data loss/disqualification, and provide states with tools to adequately evaluate public exposure risks and provide accurate public health messaging during wildfire/smoke events.

Adaptive Control Design for Uncertain Underactuated Cranes With Nonsmooth Input Nonlinearities

This article investigates the control of underactuated crane systems with unknown system parameters and nonsmooth input nonlinearities. To solve this problem, a novel filter-based adaptive control method is designed for the underactuated crane systems. First, since the backstepping control scheme cannot be directly applied to the crane systems, a group of filters is designed such that the crane systems become a class of strict-feedback nonlinear systems in each step of backstepping. Then, in order to ensure that the swing angle and position error converge to the origin, the variable transformation method is adopted. The result demonstrates that the tracking errors of the underactuated crane systems could be guaranteed to converge to a ball of origin with an arbitrarily small radius. Finally, the simulation results show that the proposed scheme is effective.

Convolutional – recurrent neural network proxy for robust optimization and closed-loop reservoir management

Production optimization under geological uncertainty is computationally expensive, as a large number of well control schedules must be evaluated over multiple geological realizations. In this work, a convolutional-recurrent neural network (CNN-RNN) proxy model is developed to predict well-by-well oil and water rates, for given time-varying well bottom-hole pressure (BHP) schedules, for each realization in an ensemble. This capability enables the estimation of the objective function and nonlinear constraint values required for robust optimization. The proxy model represents an extension of a recently developed long short-term memory (LSTM) RNN proxy designed to predict well rates for a single geomodel. A CNN is introduced here to processes permeability realizations, and this provides the initial states for the RNN. The CNN-RNN proxy is trained using simulation results for 300 different sets of BHP schedules and permeability realizations. We demonstrate proxy accuracy for oil-water flow through multiple realizations of 3D multi-Gaussian permeability models. The proxy is then incorporated into a closed-loop reservoir management (CLRM) workflow, where it is used with particle swarm optimization and a filter-based method for nonlinear constraint satisfaction. History matching is achieved using an adjoint-gradient-based procedure. The proxy model is shown to perform well in this setting for five different (synthetic) ‘true’ models. Improved net present value along with constraint satisfaction and uncertainty reduction are observed with CLRM. For the robust production optimization steps, the proxy provides O(100) runtime speedup over simulation-based optimization.

JoMIC: A joint MI-based filter feature selection method

Feature selection (FS) is a common preprocessing step of machine learning that selects informative subset of features which fuels a model to perform better during prediction or classification. It helps in the design of an intelligent and expert system used in computer vision, image processing, gene expression data analysis, intrusion detection and natural language processing. In this paper, we introduce an effective filter method called Joint Mutual Information with Class relevance (JoMIC) using multivariate Joint Mutual Information (JMI) and Mutual Information (MI). Our method considers both JMI and MI of a non selected feature with selected ones w.r.t a given class to select a feature that is highly relevant to the class but non redundant to other selected features. We compare our method with seven other filter-based methods using the machine learning classifiers viz., Logistic Regression, Support Vector Machine, K-nearest Neighbor (KNN), Decision Tree, Random Forest, Naïve Bayes, and Stochastic Gradient Descent on various datasets. Experimental results reveal that our method yields better performance in terms of accuracy, Matthew’s Correlation Coefficient (MCC) and F1-score over 16 benchmark datasets, as compared to other competent methods. The superiority of our proposed method is that it uses an effective objective function that combines both JMI and MI to choose the relevant and non redundant features.

Hybrid Meta-Heuristic based Feature Selection Mechanism for Cyber-Attack Detection in IoT-enabled Networks

Today's technologically advanced connected world is mostly reliant on the Internet of Things (IoT)-enabled smart gadgets and easy connectivity. These smart gadgets are more susceptible to malicious practices found in network traffic, which is one of the biggest challenges in cyber security domain. As a result, many systems and end-users are adversely affected by this practice. However, intrusion detection systems (IDS) are often applied to guard against cyber-attacks. Since, IDS plays a key role in detecting and preventing cyber-attacks in IoT-enabled networks, but design of an efficient and fast IDS system for cyber-attack detection is still a challenging research issue. Moreover, IDS datasets contain multiple features and to design an efficient and fast IDS, feature selection (FS) is an essential mechanism to remove the irrelevant and redundant features from large IDS datasets. Thus, this paper has proposed a hybrid feature selection scheme in which statistical test-based filter approaches such as Chi-Square (χ^2), Pearson's Correlation Coefficient (PCC), and Mutual Information (MI) are combined with a Non-Dominated Sorting Genetic Algorithm (NSGA-II)-based metaheuristic approach for optimization of features. In the proposed scheme, filter-based methods are employed to rank the features for guided population initialization in NSGA-II for faster convergence towards a solution. Performance evaluation of the proposed scheme is evaluated using the ToN-IoT dataset in terms of number of selected features and accuracy. Experimental outcomes are compared with some latest state-of-art techniques. Result analysis confirms the superior performance of the proposed scheme with minimum number of optimized features (only 13 out of 43 features) and maximum accuracy (99.48%).

Hong Kong World: Leveraging Structural Regularity for Line-based SLAM.

Manhattan and Atlanta worlds hold for the structured scenes with only vertical and horizontal dominant directions (DDs). To describe the scenes with additional sloping DDs, a mixture of independent Manhattan worlds seems plausible, but may lead to unaligned and unrelated DDs. By contrast, we propose a novel structural model called Hong Kong world. It is more general than Manhattan and Atlanta worlds since it can represent the environments with slopes, e.g., a city with hilly terrain, a house with sloping roof, and a loft apartment with staircase. Moreover, it is more compact and accurate than a mixture of independent Manhattan worlds by enforcing the orthogonality constraints between not only vertical and horizontal DDs, but also horizontal and sloping DDs. We further leverage the structural regularity of Hong Kong world for the line-based SLAM. Our SLAM method is reliable thanks to three technical novelties. First, we estimate DDs/vanishing points in Hong Kong world in a semi-searching way. We use a new consensus voting strategy for search, instead of traditional branch and bound. This method is the first one that can simultaneously determine the number of DDs, and achieve quasi-global optimality in terms of the number of inliers. Second, we compute the camera pose by exploiting the spatial relations between DDs in Hong Kong world. This method generates concise polynomials, and thus is more accurate and efficient than existing approaches designed for unstructured scenes. Third, we refine the estimated DDs in Hong Kong world by a novel filter-based method. Then we use these refined DDs to optimize the camera poses and 3D lines, leading to higher accuracy and robustness than existing optimization algorithms. In addition, we establish the first dataset of sequential images in Hong Kong world. Experiments showed that our approach outperforms state-of-the-art methods in terms of accuracy and/or efficiency.

Miniprep assisted proteomics (MAP) for rapid proteomics sample preparation.

Complete enzymatic digestion of proteins for bottom-up proteomics is substantially improved by use of detergents for denaturation and solubilization. Detergents however, are incompatible with many proteases and highly detrimental to LC-MS/MS. Recently; filter-based methods have seen wide use due to their capacity to remove detergents and harmful reagents prior to digestion and mass spectrometric analysis. We hypothesized that non-specific protein binding to negatively charged silica-based filters would be enhanced by addition of lyotropic salts, similar to DNA purification. We sought to exploit these interactions and investigate if low-cost DNA purification spin-filters, 'Minipreps,' efficiently and reproducibly bind proteins for digestion and LC-MS/MS analysis. We propose a new method, Miniprep Assisted Proteomics (MAP), for sample preparation. We demonstrate binding capacity, performance, recovery and identification rates for proteins and whole-cell lysates using MAP. MAP recovered equivalent or greater protein yields from 0.5-50 μg analyses benchmarked against commercial trapping preparations. Nano UHPLC-MS/MS proteome profiling of lysates of Escherichia coli had 99.3% overlap vs. existing approaches and reproducibility of replicate minipreps was 98.8% at the 1% FDR protein level. Label Free Quantitative proteomics was performed and 91.2% of quantified proteins had a %CV <20% (2044/2241). Miniprep Assisted Proteomics can be performed in minutes, shows low variability, high recovery and proteome depth. This suggests a significant role for adventitious binding in developing new proteomics sample preparation techniques. MAP represents an efficient, ultra-low-cost alternative for sample preparation in a commercially obtainable device that costs ∼$0.50 (USD) per miniprep.

A Target-Aware Fusion Framework for Infrared and Visible Images

Infrared and visible image fusion aims to obtain an image that can retain the prominent infrared target and the detailed texture information from the source images. Most scale filter-based decomposition methods mainly attempt to extract more detailed features by increasing decomposition layers, and they ignore fully considering the intrinsic properties of the original images and the influence of noises. To solve this issue and obtain better fusion & detection performance, this paper proposes a fusion method via a globalet filter and detail enhancement model to construct a target-aware fusion framework. The globalet filter is first comprehensively considered from three perspectives based on global point-to-point optimization: target brightness, removal of gradients of large-small scales (noise reduction), and preservation of texture details. Mathematically, three limited error measure equations are constructed between the target output and the source image in the shape of the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$L_{2}$ </tex-math></inline-formula> -norm and the first-order derivative difference. Next, a weighted average operator and a detail enhancement model are proposed to guide the corresponding sub-layers. This model creates connections between the detail layers and the input images so that the fused detail layer contains the best pixel region from these images to the greatest extent possible after the “maximum absolute” rule. As a result, the fused image is reconstructed by adding the previously obtained sub-images. Extensive experimental results demonstrate that our method outperforms state-of-the-art fusion methods, particularly in highlighting infrared targets, preserving substantial details, and producing average detection accuracy of exceeds 98%.

An Approach Combining Feature Selection with Machine Learning Techniques for Prediction Reliability and Accuracy in Hepatitis Diagnosis

Viral hepatitis is the inflammation and damage of liver cells due to infection. There are different causes of hepatitis, but the symptoms can be similar. As of 30 June 2022, 473 cases of acute hepatitis of unknown aetiology among children aged 16 years and under have been reported from the World Health Organization European Region. Simply more than (56.7%) of these cases have been reported from the UK. The majority (76.1%) of reported cases are five years old or younger. In this review, some machine learning order procedures are applied to Hepatitis informational index procured from UCI AI Storehouse. Naïve Bayes Classifier, Logistic Regression, and J48 Decision Tree are utilized as grouping algorithms and they have been contrasted agreeing with the filter-based selection method. For channel-based feature determination, Cfs Subset Eval, Data Gain Attribute Eval, and Principal Components have been utilized and their presentation of them is assessed as far as accuracy, review, F-Measure, and ROC Region. Among the pre-owned classification algorithms, Naïve Bayes Classifier has higher classification accuracy on the data set than the others with applied and nonapplied filter-based feature choices. Additionally, we proclaim that the best filter-based feature choice is Principal Components because of the highest classification precision for hepatitis patients.

The approach to adjusting commercial PM2.5 sensors with a filter-based gravimetric method

The measurements of temporal change to indoor contaminant concentrations are critical to understanding pollution characteristics. As commercial sensors are becoming increasingly commonplace, concentration accuracy is still a critical issue. The most common methods for measuring indoor particulate pollutants based on filter-based gravimetric methods. However, the gravimetric method is expensive, time-consuming, and often provides little temporal information. More and more commercial sensors are utilized to collect larger and temporal information about indoor air pollutants. Nevertheless, limited data support the accuracy of commercial sensors so far. Thus, this study aims to evaluate the performance of commercial sensors. PM2.5 were collected for 30 days by personal environmental monitors with an airflow of 2 L/min on 37-mm Teflon filters and commercial sensors, simultaneously in a three-story house. Moreover, the intra-sensor comparison was conducted for 24 hours by the resolution in 1 minute. Finally, the linear regression model was built to adjust commercial sensors. The intra-sensor comparison results revealed that 24 hours average coefficient of variation (CV value) of PM2.5 in this study were under 10% and the R2 of the adjusted equation was 0.9394. We provide an accurate concentration of commercial sensors to estimate the association between pollutants exposure and health.

Complete genome sequence of "Candidatus Phytoplasma sacchari" obtained using a filter-based DNA enrichment method and Nanopore sequencing.

Phytoplasmas are phloem-limited plant pathogens, such as sugarcane white leaf (SCWL) phytoplasma, which are responsible for heavy economic losses to the sugarcane industry. Characterization of phytoplasmas has been limited because they cannot be cultured in vitro. However, with the advent of genome sequencing, different aspects of phytoplasmas are being investigated. In this study, we developed a DNA enrichment method for sugarcane white leaf (SCWL) phytoplasma, evaluated the effect of DNA enrichment via Illumina sequencing technologies, and utilized Illumina and Nanopore sequencing technologies to obtain the complete genome sequence of the "Candidatus Phytoplasma sacchari" isolate SCWL1 that is associated with sugarcane white leaf in China. Illumina sequencing analysis elucidated that only 1.21% of the sequencing reads from total leaf DNA were mapped to the SCWL1 genome, whereas 40.97% of the sequencing reads from the enriched DNA were mapped to the SCWL1 genome. The genome of isolate SCWL1 consists of a 538,951 bp and 2976 bp long circular chromosome and plasmid, respectively. We identified 459 protein-encoding genes, 2 complete 5S-23S-16S rRNA gene operons, 27 tRNA genes, and an incomplete potential mobile unit (PMU) in the circular chromosome. Phylogenetic analyses and average nucleotide identity (ANI) and digital DNA-DNA hybridization (dDDH) values based on the sequenced genome revealed that SCWL phytoplasma and sugarcane grassy shoot (SCGS) phytoplasma belonged to the same phytoplasma species. This study provides a genomic DNA enrichment method for phytoplasma sequencing. Moreover, we report the first complete genome of a "Ca. Phytoplasma sacchari" isolate, thus contributing to future studies on the evolutionary relationships and pathogenic mechanisms of "Ca. Phytoplasma sacchari" isolates.

A Novel Approach to Increase the Efficiency of Filter-Based Feature Selection Methods in High-Dimensional Datasets With Strong Correlation Structure

A Novel Approach to Increase the Efficiency of Filter-Based Feature Selection Methods in High-Dimensional Datasets With Strong Correlation Structure

Motion profile and control design of fast steering mirror for backscanning step and stare imaging.

Fast steering mirrors (FSMs) have been used for decades to improve the performance of electro-optical imaging systems, such as airborne imaging systems and space-based optical surveillance systems. With the advantage of increasing the accuracy of image motion compensation and the efficiency of scanning imaging, backscanning step-and-stare imaging has become the main approach to realizing wide-area surveillance for airborne imaging systems. According to the operating mode and motion profile of the FSM in the imaging system, a combined optimized profile (COP) is designed to avoid abrupt changes in the velocity and acceleration of the FSM. The angular position sensor based on a four-quadrant detector is used in FSMs to expand the measuring range and cut the cost at the expense of larger measurement noise. Combining a Kalman filter with a disturbance observer and zero-phase error tracking control, a control method is proposed to improve the control precision and bandwidth while suppressing measurement noise. Simulation and experimental results show that the profile designed by COP is smooth enough to meet the special requirements of FSM's backscanning image motion compensation and that the Kalman filter-based FSM control method can significantly improve the control accuracy.

3D Mapping for a Large Crane Using Rotating 2D-Lidar and IMU Attached to the Crane Boom

This paper describes a novel method for large-scale 3D mapping for construction cranes with an arbitrary motion of the sensor system (2D lidar and IMU) attached to the crane boom. A heavy lidar with a slowly rotating base is needed to make a large-scale map both vertically and horizontally for cranes. This sensor configuration and mapping conditions entail handling each 2D scan separately, making it difficult to adopt existing rotating 2D lidar-based methods that construct a virtual 3D scan from a set of 2D scans. In the proposed method, we introduce a complementary filter with moving average filtering for lidar pose estimation, which is more robust to severe vibration than Kalman filter-based methods. As there are only a small amount of overlaps between 2D lidar scans, we propose a map correction method based on a pose graph optimization with planar environmental constraints. We evaluate the proposed method in a simulation and a small-sized real environment and compare it with one of the state-of-the-art methods. The evaluation results reveal that the proposed method can accurately estimate the sensor poses, thereby generating a high-quality, large-scale 3D point cloud map.

PROMETHEE filter-based method for microarray gene expression data

Gene expression datasets have been successfully applied for a variety of purposes, including cancer classification. The challenges faced in developing effective classifiers for expression datasets are high dimensionality and over-fitting. Gene selection is an effective and efficient method to overcome these challenges and improve the predictive accuracy of a classifier. Based on PROMETHEE, this paper introduces a multi-filter ensemble approach by integrating the results of two potential filters namely MaCΨ-filter and PCRWG-filter to pre-select the most informative genes. Experiments were conducted on nine microarray datasets to demonstrate the performance of the proposed method.

Selection of regional determinants of economic security

Subject. The article addresses the selection of informative and significant regional determinants of economic security. Objectives. The aim is to explore statistical and econometric tools to select regional determinants, to build economic security classifiers, using the machine learning algorithms. Methods. The determinants are selected, using the filter-based methods (on the basis of Chi-square, F-criterion, information content criterion), sequential and recursive selection in the opposite direction, exhaustive sampling and selection based on the "random forest" classifier. The above algorithms are implemented in the Scikit-learn and MLxtend libraries. The analytical environment Jupyter Notebook is used as a system providing the execution of feature selection algorithms. Results. I formed a training sample of 90% of the total set of regional economic determinants for 82 Russian regions. The determinants were ranked by significance. Logistic loss and model accuracy were set as a model evaluation criterion. The accuracy of created classifiers was between 78% and 93%. The paper establishes the significance of categorical variables that determine whether a particular region belongs to the resource, investment, or regional profitability cluster. The constructed heat map enabled to establish the degree of linear relationship between the level of economic security of the region and regional determinants. Conclusions. The most significant group of panel indicators to determine the class of economic security is the resource security of regions. The level of resource endowment depends on the volume of consumption and the share of manufactured goods in the GRP. To build a model to classify regions by the economic security level, five features are sufficient; this circumstance was established by automatic set of determinants, using the recursive feature elimination algorithm with cross-validation (RFECV). The implemented wrapping methods of feature selection demonstrated that significant indicators in the model are budget expenditures and the amount of debt per capita, investment risk and net financial result. The best quality of the classifier is provided in the course of an exhaustive set of features, using the K-Nearest Neighbors algorithm.

ATSUKF-Based Actuator Health Assessment Method for Quad-Copter Unmanned Aerial Vehicles

The actuator, which generally consists of motors, electrical regulations, and propellers, is the key component of the quadrotor Unmanned Aerial Vehicle. During the operation of the UAV, actuators are prone to degrade performance and even cause serious failure, which affects the service quality and flight safety of Unmanned Aerial Vehicles. Therefore, timely and accurate monitoring and evaluation of the health condition of actuators is of great significance to ensure the mission reliability of UAVs. This paper proposes an Adaptive Two-stage Unscented Kalman Filter-based actuator health assessment method for Quadcopter Unmanned Aerial Vehicles. Firstly, a state space equation is established based on dynamic analysis to characterize the degradation mechanism of the actuator. Then, by modifying the Two-stage Unscented Kalman Filter algorithm, the Adaptive Two-stage Unscented Kalman Filter algorithm is constructed by combining the filter divergence criterion and the covariance matching technique to implement the health assessment of actuators. Finally, experiments are carried out for different degradation scenarios to verify the effectiveness of the proposed method.

Comparing Practical Spectral Imaging Methods for Cultural Heritage Studio Photography

Two practical methods for implementing spectral imaging within the framework of museum studio photography were investigated. Imaging was carried out using a consumer RGB digital camera paired with either (1) colored glass filters and a broadband source or (2) optimized multichannel LED illumination, yielding five or six spectral image bands, respectively. Color targets were used to develop and verify profiles for transforming between the multiband camera signals and final color managed images. The filter-based and LED-based profiles were assessed quantitatively for color accuracy using color difference statistics, and several paintings were imaged and rendered using the profiles as a visual demonstration of the differences. While both were superior to conventional RGB imaging, the LED-based method outperformed the filter-based method for accurate reproduction of independent data. This supplements practicality and cost considerations that are informing the development of accessible spectral imaging strategies for highly color accurate museum studio photography.

A Kalman filter-based online fractional cycle bias determination method for real-time ambiguity-fixing GPS satellite clock estimation

Integer ambiguity resolution (IAR) is one of the most powerful methods to improve the precision and the convergence of estimating real-time satellite clocks, which is an indispensable prerequisite to support the real-time precise point position (PPP) service. In this study, a Kalman filter-based online fractional cycle bias (FCB) determination method is proposed to facilitate the real-time IAR for satellite clock estimation. A real-time recursive algorithm for smoothing the average float wide-lane (WL) ambiguity is developed to mitigate the impact of pseudorange noise. Quality control procedures are established throughout the entire FCB filter process to remove low-quality observations and mitigate the impact of incorrect integer ambiguities. This method is designed as an embedded module in the clock estimation procedure to fix both integer WL and narrow-lane (NL) ambiguities in an epoch-by-epoch way for recovering the integer property of the undifferenced ionosphere-free (IF) ambiguities in real-time. The performance of the FCB filter and the ambiguity-fixing clock estimation is validated with a GPS network from the international GNSS service (IGS). Experiments show that the WL and NL FCB estimation accuracy reaches 0.004 cycles and 0.080 cycles for GPS satellites. The WL and NL residuals fit the normal distribution well after several minutes and 1.5 h, respectively, making the ambiguity-fixing clocks converge rapidly in about two hours. The online FCB filter method improves real-time ambiguity-fixing clock estimation accuracy significantly. Compared to the ambiguity-float solution, the STD of the ambiguity-fixing real-time satellite clocks is averagely improved by 31.0 % to 0.040 ns with the ultra-rapid orbits. The ambiguity-fixing satellite clocks can be applied in the real-time kinematic PPP ambiguity resolution, and the positioning accuracy is improved by 24.1 % to 51.7 % compared to that of using the ambiguity-float clock products.

A hybrid machine learning technique for feature optimization in object-based classification of debris-covered glaciers

Object-based features like spectral, topographic, and textural are supportive to determine debris-covered glacier classes. The original feature space includes relevant and irrelevant features. The inclusion of all these features increases the complexity and renders the classifier’s performance. Therefore, feature space optimization is requisite for the classification process. Previous studies have shown a rigorous exercise in manually selecting the best combination of features to define the target class and proven to be a time-consuming task. The present study proposed a hybrid feature selection technique to automate the selection of the best suitable features. This study aimed to reduce the classifier’s complexity and enhance the performance of the classification model. Relief-F and Pearson Correlation filter-based feature selection methods ranked features according to the relevance and filtered out irrelevant or less important features based on the defined condition. Later, the hybrid model selected the common features to get an optimal feature set. The proposed hybrid model was tested on Landsat 8 images of debris-covered glaciers in Central Karakoram Range and validated with present glacier inventories. The results showed that the classification accuracy of the proposed hybrid feature selection model with a Decision Tree classifier is 99.82%, which is better than the classification results obtained using other mapping techniques. In addition, the hybrid feature selection technique has sped up the process of classification by reducing the number of features by 77% without compromising the classification accuracy.