Central Pixel Research Articles

Adaptive center pixel selection strategy in Local Binary Pattern for texture classification

Local Binary Pattern (LBP) is widely used in texture classification because of its powerful capability to extract texture features of a center pixel. However, LBP has three main drawbacks: (1) limited by the low resolution of imaging device, the quality of texture image is degraded, and some real existing pixels with more texture information are unavoidably lost. (2) Center pixel gc is the most important factor to extract correct LBP pattern. However, by far LBP and its variants do not show any solutions to enhance the robustness of center pixel gc. (3) Some LBP patterns include important texture microstructures, but they are ignored by uniform patterns. At the same time, some uniform patterns can be corrupted by noise and misclassified into non-uniform patterns. These LBP patterns therefore all lost their discrimination capability. In order to overcome these three disadvantages, in this paper, we propose a novel adaptive center pixel selection (ACPS) strategy.Inspired by image super-resolution techniques, ACPS firstly applies the interpolation method to recover the lost real existing pixels and generate the center pixel candidates with more texture information. Then, the gradient information is used to obtain the edge image aiming to find the non-uniform patterns at edge points which may contain complicated texture microstructures. After generating the center pixel candidates and edge image, we introduce ACPS strategy into the LBP framework. By adaptively selecting the optimal center pixel from all center pixel candidates, one non-uniform pattern at the edge point can be possibly recovered to the uniform pattern, and regain its discrimination power. It is worth noting that any other LBP variants can also employ the ACPS strategy to more effectively extract its texture features. By observing the experimental results on representative texture databases of Outex, UIUC, CUReT, XU_HR, ALOT and KTHTIPS2b after introducing the ACPS strategy into LBP and its variants of LTP, CLBP, BRINT, CRDP, FbLBP, and CJLBP, the texture classification performances can be significantly improved.

Facial expression analysis using local directional stigma mean patterns and convolutional neural networks

This paper represents automatic facial expression analysis method named Local Directional Stigma Mean Patterns (LDSMP) for automatic facial expression analysis and image retrieval using content based facial expression image retrieval and CNN. The traditional local patterns such as Local Binary Patterns (LBP) and Local Ternary Patterns (LTP) are applied for face recognition and expression analysis, calculated using relationship between the center pixel and neighboring pixels. The proposed method calculates the eight directional difference values then divided into the three ranges based on threshold values. Thus, the values are substituted with basic three positive values (+3, +2, +1) and three negative values (-3, -2, -1) to get more sensitive information from an image rather than aforementioned methods. The threshold can be select either static which is selected by user or dynamic is evaluated from image itself and supports to improve the efficiency. The performance of the proposed method is further improved by giving this patterns as input to the Convolutional Neural Networks (CNN) and compared with the existing methods LBP, LTP, and Directional Binary Code (DBC) in terms of Average Precision (AP), Average Recall (AR), and Average Retrieval Rate (ARR) using standard databases COREL 10K (DB1) and JAFFE (The Japanese Female Facial Expression) (DB2) and Extended Cohn-Kanade (CK +) (DB3) dataset.

Machine learning application in Glioma classification: review and comparison analysis

This paper simply presents a state of the art survey among the machine learning based approaches for the Glioma classification. As Glioma classification is a very challenging task in the field of medical science and this task is well addressed and taken by the fraternity of machine learning experts, who are working day and night to devise automated approaches that can automate this whole process of Glioma accurate classification from the various medical imaging modalities like Magnetic resonance imaging (MRI), Computed tomography (CT) etc. Although present machine learning techniques offers an opportunity to come up with a highly accurate and automated Glioma classification approach, by performing fusion among the various medical imaging modalities as well as utilizing the various features derived from the multi-modality medical imaging data. This paper also proposed an efficient and accurate automated approach of Glioma classification for the comparison analysis. This proposed approach is based on the use of hybrid ensemble learning model and hybrid feature extraction method, which relies on the Discrete wavelet Decomposition (DWD), Central pixel Neighbourhood Binary pattern (CNBP) and GLRLM (Gray level run length Matrix) methods in order to classify the Glioma (type of mostly diagnosed brain tumors) into Low grade Glioma and High grade Glioma from the fused MRI sequences. Improved eXtreme Gradient Boosting classifier is the hybrid ensemble learning model, which is used in this paper for the first time along with the proposed hybrid texture feature extraction method. Further this proposed approach is compared with the already existing state of the art approaches, which are based on the various machine learning classifiers like Support vector machine (SVM), K-Nearest neighbor (KNN), Naive Bayes (NB) etc. and conventional feature extraction methods in order to present a comprehensive and practical comparison study. The proposed approach is evaluated on the balanced large size local dataset consisting of MRI images of low and high grade Glioma collected from the various MRI centers located in Madhya Pradesh, India as well as on the popular global datasets like, BRATS 2013 and BRATS 2015 with various MRI fusion combinations (T1 + T1C + T2 + Flair, T1 + T1C + T2, T1 + T1C + Flair, T1C + T2 + Flair etc.). The proposed approach employing Improved eXtreme Gradient Boosting ensemble model offers highest accuracy of above 90% on the local dataset with the fusion of T1C + T2 + Flair MRI sequences.

Optimization of Edge Extraction Algorithm for Objects in Complex Background

In the field of artificial intelligence, machine vision is expected because of its low cost and easy popularization. Feature extraction is one of the core steps of machine vision. When we use Canny algorithm, which extracts edges by gray level change, we cannot directly separate the object from the background. The edge information of the object will also be lost when the Gaussian filter smoothes the image. The research proposes an improved edge extraction algorithm based on Canny algorithm. We use the image difference method and object envelopment method to separate the object from the background, so the data to be processed is greatly reduced and the speed and accuracy of the algorithm are improved. Then we use bilateral filtering instead of Gaussian filtering, which increases the weight of gray difference on the basis of Gaussian filtering, so we can to retain more edge information. Finally, we improve the gradient operator, by determining the gradient operator coefficient according to the inverse ratio of the Euclidean distance between the pixel and the center pixel. The average performance of the new gradient operator in the center of the picture is better than that of Sobel operator. Experimental simulation shows that the algorithm has good detection accuracy. Compared with traditional algorithms, the improved algorithm is not affected by complex backgrounds, and reduce the influence of light. The improved algorithm has clear edges in the center area of the image.

An adaptive atmospheric correction algorithm for the effective adjacency effect correction of submeter-scale spatial resolution optical satellite images: Application to a WorldView-3 panchromatic image

The adjacency effect can blur and reduce the contrast of satellite images. For low visibility, the adjacency effect has a crucial impact on submeter-scale spatial resolution optical (SM) satellite images. Therefore, SM satellite images are preprocessed with an atmospheric correction before performing visual interpretation and quantitative research. This atmospheric correction includes the adjacency effect (expressed by the average background reflectance, ABR) correction and atmospheric intrinsic reflectance correction. ABR is related to the spatial distance from the central pixel to its surrounding pixels (SDCS) and is affected by the reflectance difference between the central pixel and its surrounding pixels (RDCS). Existing methods to calculate ABR do not include the effect of RDCS. Therefore, in this study, we use equivalent average background reflectance (EABR) instead of ABR used in the 6S model. EABR considers the effects of the strengths of the aerosol optical depth (AOD), SDCS, and RDCS on the adjacency effect. Compared with ABR, EABR can better represent the adjacency effect in an SM satellite image. We develop an adaptive atmospheric correction algorithm (called adaptive-AC) based on EABR. The adaptive characteristic of adaptive-AC is that it can adjust the contribution factor (or weighted value) of the background pixel to the adjacency effect according to the SDCS and RDCS. The application of adaptive-AC to a panchromatic band image of WorldView-3 (WV-3) reveals that adaptive-AC can significantly improve image quality, with the reflectance of pixels in the corrected image processed by adaptive-AC being close to the actual reflectance. Comparison with the results of the atmospheric correction algorithm in the 6S model (6S-AC) indicates that adaptive-AC is more suitable for SM satellite image preprocessing than 6S-AC approaches. There is no clear theory or formula available to calculate the exact value of the adjacency effect range (Rad). Therefore, different Rad values are used in applying the adaptive-AC and 6S-AC. The results show that the larger is the Rad used to construct the adjacency effect model, the more satisfactory is the atmospheric correction.

Variant center-symmetric census transform for real-time stereo vision architecture on chip

Stereo vision is one of the most crucial operations in many computer vision applications, and stereo matching is its most important step. In recent years, stereo matching has developed in the direction of increasing accuracy and higher computational speed, and the real-time stereo matching algorithm based on hardware architecture has been increasingly emphasized due to its applicability in embedded systems. The most frequently used method on FPGA is the census transform (CT) because of its simple structure and easy parallelization and the high quality of the generated disparity maps, but CT has the drawbacks of mismatches in some regions and dependence on a central pixel. In this paper, an improved CT-based semi-global stereo matching algorithm with pipeline and parallel operation based on FPGA is proposed in order to increase matching accuracy in specific regions while satisfying real-time constraints. In the matching cost step, pixels are divided into two parts, with the two methods calculated to generate the bit-vector feature. A four-path semi-global matching algorithm with twice aggregate is proposed in the cost aggregation stage. The left–right check, mismatching point filling, and the median filter to enhance the final disparity map are also required. The novel algorithm is evaluated on the Middleburry benchmark and implemented on Xilinx ZYNQ-7000 SoCs, which results in a throughput of 640 × 480/60 fps, with 64 disparity levels at 100 MHz. Compared with the related work, we improve the average accuracy by 1.61%, making our approach suitable for real-time embedded systems.

A new methodology for offshore wind speed assessment integrating Sentinel-1, ERA-Interim and in-situ measurement

Offshore Wind (OW) speed assessment is a key aspect for the development of new wind farms at sea. Satellites can be used to globally obtain ocean and sea distribution, compensating limited in-situ measurements. In this study, a new methodology to estimate the wind’s speed potential is here proposed. Preliminary, Sentinel-1 (S-1) images have been analyzed by means of the Sentinel Application Platform (SNAP) software, extrapolating wind speed data for each cell pixel size of a testing area. Then GIS (Geographic Information System) software has been used to map wind data and find the best pixel location comparing these data with in-situ data. Furthermore, wind speed has been analyzed using the ERA-Interim reanalysis dataset for areas within 11 km and 40 km from the Lillgrund OW farm in the Baltic Sea to better understand wind regimes. Finally, wind speed parameters obtained by S-1 in Sea Surface Water (SSW) with the 10 m standard high have been compared with wind speed recorded by Supervisory Control and Data Acquisition (SCADA) systems of two turbine using wind profile formula. Obtained results show the comparison accuracy of wind speed assessment for each center of the pixels by S-1 satellite images and in-situ (SCADA) measurements. Data actually depends on the distance between the selected center pixel and the location of the turbines. The obtained wind speed differences (0.26 m/s - RMSE = 1.38 and 0.92 m/s - RMSE = 1.82) pinpointed the direct effect of the distance between the selected pixel center and the in-situ measurements location in the S-1 imagery for wind speed potential assessment. Obtained results proved an improvement of the OW assessment accuracy using multiple satellite observations, demonstrating that SAR wind maps can support OW speed sites assessment by introducing observations in different phases of an OW farm project.

Mean distance local binary pattern: a novel technique for color and texture image retrieval for liver ultrasound images

A rapid growth in medical ultrasound database makes it difficult for medical practitioners to manage and search relevant data with good efficiency. Hence, a novel image retrieval technique using Mean Distance Local Binary Pattern (Mean Distance LBP) has been proposed for content-based image retrieval. The conventional local binary pattern (LBP) converts every pixel of image into a binary pattern based on their relationship with neighbourhood pixels. The proposed feature descriptor differs from local binary pattern as it transforms the mutual relationship of all neighbouring pixels in a binary pattern based on their standard deviation templates as well as Euclidean distance from the center pixel. Color feature and Gray Level Co-occurrence Matrix have also been used in this work. To prove the excellence of the proposed method, experiments have been conducted on two different databases of natural images and face images. Further, the method is applied on real time ultrasound database for retrieval of liver images from a set of ultrasound images of various organs. The performance has been observed using well-known evaluation measures, precision and recall, and compared with some state-of-art local patterns. Comparison shows a significant improvement in the proposed method over existing methods.

Analyzing the Ecological Environment of Mining Area by Using Moving Window Remote Sensing Ecological Index

Objectives Mine environmental problems are intensifying, and the environmental monitoring and evaluation of the mining area is an indispensable part of mine management. Previous researches have obtained effective results by evaluating the mining ecological environment through the remote sensing ecological index (RSEI). However, the calculation does not take into account that the environmental influence is regional under natural conditions of RSEI, and the geo-graphical location of the mining area has different effects on its surrounding ecological environment. The main purpose of this research is to propose a new remote sensing ecological index that is more in line with the ecological response mechanism based on the characteristics of the ecological environment of the mining area. Methods Aiming at the complex research area of large area feature type, this paper presents an improved ecological index named moving window-based remote sensing ecological index (MW-RSEI) which based on RSEI and moving window evaluation unit. MW-RSEI divides the ecological environment into four influencing factors: Greenness, wetness, dryness and heat. Each pixel in the study area is taken as the research object, and the extended window of the pixel is taken as the calculation range of each factor. The experiment assigns weight to the center pixel of each window through principal component analysis. Results The MW-RSEI and RSEI models were used to calculate in the study area, the results show that the mean value of MW-RSEI and RSEI in the study area were 0.522 and 0.459, the greenness and wetness had positive effect on the environment, and the dryness and heat had negative effects. Based on the MW-RSEI model, mine geomorphology destroys the regional ecological environment grade is worst and poor accounted for 85.40% of the overall area. MW-RSEI in line with the first law of geography. Its results are consistent with RSEI, and it shows more gradient information of the ecological environment around the mining area, which confirms the theory of dissipative structure in the ecological environment. Conclusions MW-RSEI refines the local characteristics of the ecological environment, and it can provide an effective basis for ecological environment monitoring and evaluation in mining areas.

Research on Noise Reduction and Enhancement Algorithm of Girth Weld Image

In order to eliminate the salt pepper and Gaussian mixed noise in X-ray weld image, the extreme value characteristics of salt and pepper noise are used to separate the mixed noise, and the non local mean filtering algorithm is used to denoise it. Because the smoothness of the exponential weighted kernel function is too large, it is easy to cause the image details fuzzy, so the cosine coefficient based on the function is adopted. An improved non local mean image denoising algorithm is designed by using weighted Gaussian kernel function. The experimental results show that the new algorithm reduces the noise and retains the details of the original image, and the peak signal-to-noise ratio is increased by 1.5 dB. An adaptive salt and pepper noise elimination algorithm is proposed, which can automatically adjust the filtering window to identify the noise probability. Firstly, the median filter is applied to the image, and the filtering results are compared with the pre filtering results to get the noise points. Then the weighted average of the middle three groups of data under each filtering window is used to estimate the image noise probability. Before filtering, the obvious noise points are removed by threshold method, and then the central pixel is estimated by the reciprocal square of the distance from the center pixel of the window. Finally, according to Takagi Sugeno (T-S) fuzzy rules, the output estimates of different models are fused by using noise probability. Experimental results show that the algorithm has the ability of automatic noise estimation and adaptive window adjustment. After filtering, the standard mean square deviation can be reduced by more than 20%, and the speed can be increased more than twice. In the enhancement part, a nonlinear image enhancement method is proposed, which can adjust the parameters adaptively and enhance the weld area automatically instead of the background area. The enhancement effect achieves the best personal visual effect. Compared with the traditional method, the enhancement effect is better and more in line with the needs of industrial field.

Weber’s law based multi-level convolution correlation features for image retrieval

Weber’s law reveals the relationship between human perception and perceptual stimuli. Inspired by the theory, this paper designs a multi-level convolution correlation feature statistic method for image retrieval. Firstly, the difference between a central pixel and its neighbors is described by Weber’s law through computing the differential excitation of image. Then, a multi-level saliency map is obtained by binary transformation and convolution operation. Thirdly, to exploit spatial correlation information of the image, a pixels pair-wise correlation and hierarchy statistic model is constructed. Finally, all intermediate features are concatenated into one histogram, which includes salient color and texture features. Extensive experiments demonstrate the proposed method of this paper has excellent performance.

Local Neighbourhood Edge Responsive Image Descriptor for Texture Classification Using Gaussian Mutated JAYA Optimization Algorithm

Local feature descriptor plays a significant role in texture classification. However, in the traditional local binary pattern method, image pixels are converted into a binary pattern based on the relationship between center and neighborhood pixels. This paper introduces a novel feature extraction method named local neighbourhood edge responsive binary pattern to extract and categorize reliable texture features from images for achieving an automated classification of different types of ocean bottom sediments. Initially, the basic magnitude variance pixel values are derived depending on an odd and even pixel value of a 3 × 3 image patch. Further, the edge information is extracted using the local directional pattern method from all images. The edge response of the image is obtained using a kirsch mask in all eight directions. The encoding condition is then applied to both the local intensity and the edge information to create a unique descriptor value. Finally, a new learning algorithm called GMJAYA-ELM combines the Gaussian mutated JAYA (GMJAYA) with an extreme learning machine (ELM) for texture classification. The GMJAYA is used to optimize the input weights and hidden biases of single-hidden-layer feed-forward neural networks. Experimental findings indicate that the suggested method yields better accuracy and sensitivity efficiency among various groups. The proposed algorithm is tested by contrasting outcomes with standard learning schemes such as PSO-ELM, GA-ELM, ABC-ELM, Birdswarm-ELM, and JAYA-ELM, suggesting the dominance of GMJAYA-ELM.

Accurate subpixel shifting of point spread functions.

The localization of emitters requires accurate subpixel shifting of point spread function (PSF) models. However, the PSF recorded by the camera is not the true PSF of the system due to integration across finite pixels. These errors can be propagated during the shifting process, causing systematic biases in the registration or localization process. This letter proposes a set of filter kernels that, when convolved with the image, accurately shifts it by an arbitrary subpixel shift. Each pixel in the filter is represented by a two-dimensional polynomial function of the possible x and y shift values. These filters are effective; when tested on three different PSFs, they reduced errors by a factor of 20 or more over PSF models evaluated at the pixel center.

Infrared small target detection using tensor based least mean square

Infrared small target detection is an important topic in infrared image processing and pattern recognition. It plays an important role in reconnaissance, early warning system, aircraft tracking and missile guidance. Because of the low signal-to-noise ratio, small target size, obvious shape structure and texture information available, infrared small target detection is a very difficult task. Inspired by the observation that infrared small targets usually exist in image background and the image has matrix structure, we propose to observe the infrared small target in main view and develop a Tensor based least mean square (TLMS) method to detect infrared small target. In TLMS, the neighborhood of the target is firstly utilized to predict the gray value of the central pixel. The predicted background image and the difference image are derived from the principle of mean square minimum error. Finally, the target is obtained by performing the adaptive thresholding. We conduct the experiments on four datasets to compare the performance of TLMS with Max-median/Max-mean, Top-hat, left-TDLMS and right-TDLMS. The experimental results show that TLMS is an effective infrared small target detection method which can well suppress the background and highlight the infrared small targets.

Image-based rock typing using local homogeneity filter and Chan-Vese model

Image-based rock typing is carried out to classify an image of the heterogeneous rock sample into different rock types where each rock type can be treated as a homogeneous porous medium. In this study, we propose an innovative method for rock typing of the heterogeneous rock sample via three steps. First, the target image, a segmented binary image with two phases of pore and solid, is consecutively inputted into two filters of a local homogeneity filter and an average filter to increase the contrast between different rock types and decrease the contrast within each single rock type. Second, Chan-Vese model is applied to classify the filtered image into different rock types. Third, a thresholding is used to remove the particles, which are treated as noisy particles, smaller than a given preset size. The main idea of the local homogeneity filtering introduced in this study is undertaken by counting the number of pixels that possess the same phases as the center pixel within a 3 × 3 pixels neighborhood. This process is carried out iteratively, which means the previously estimated pixel will be used in the estimation of its neighbor unprocessed pixels. We demonstrate the application of the proposed method in several heterogeneous images and present good performance.

Hyperspectral Image Classification With Multiattention Fusion Network

Hyperspectral image (HSI) has hundreds of continuous bands that contain a lot of redundant information. Besides, a spatial patch of a hyperspectral cube often contains some pixels different from the center pixel category, which are usually called interference pixels. The existence of such interference pixels has a negative effect on extracting more discriminative information. Therefore, in this letter, a multiattention fusion network (MAFN) for HSI classification is proposed. Compared with the current state-of-the-art methods, MAFN uses band attention module (BAM) and spatial attention module (SAM), respectively, to alleviate the influence of redundant bands and interfering pixels. In this way, MAFN realizes feature reuse and obtains complementary information from different levels by combining multiattention and multilevel fusion mechanisms, which can extract more representative features. Experiments were conducted on two public HSI data sets to demonstrate the effectiveness of MAFN. Our source code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Li-ZK/MAFN-2021</uri> .

Precision Weak Gravitational Lensing Using Velocity Fields: Fisher Matrix Analysis

Abstract Weak gravitational lensing measurements based on photometry are limited by shape noise: the variance in the unknown unlensed orientations of the source galaxies. If the source is a disk galaxy with a well-ordered velocity field, however, velocity field data can support simultaneous inference of the shear, inclination, and position angle, virtually eliminating shape noise. We use the Fisher information matrix formalism to forecast the precision of this method in the idealized case of a perfectly ordered velocity field defined on an infinitesimally thin disk. For nearly face-on targets, one shear component, γ ×, can be constrained to , where I 0 is the signal-to-noise ratio of the central intensity pixel and n pix is the number of pixels across a diameter enclosing 80% of the light. This precision degrades with inclination angle by a factor of 3 by i = 50°. The uncertainty on the other shear component, γ +, is about 1.5 (7) times larger than the γ × uncertainty for targets at i = 10° (50°). For an arbitrary galaxy position angle on the sky, these forecasts apply not to γ + and γ × as defined on the sky but to two eigenvectors in (γ +, γ ×, μ) space, where μ is the magnification. We also forecast the potential of less expensive partial observations of the velocity field, such as slit spectroscopy. We conclude by outlining some ways in which real galaxies depart from our idealized model and thus create random or systematic uncertainties not captured here. In particular, our forecast γ × precision is currently limited only by the data quality rather than scatter in galaxy properties because the relevant type of scatter has yet to be measured.

Wood species automatic identification from wood core images with a residual convolutional neural network

This paper tackles the problem of automatic tree species identification from scanned images of wood cores. A convolutional neural network with residual connections is proposed to perform this task. The model is applied to consecutive image patches following the sliding window strategy to recognize a patch central pixel’s membership. It then decides about the resulting tree species via a majority voting. The model’s performance was assessed concerning a dataset of 312 wood core images representing 14 European tree species, including both conifer and angiosperm (ring-porous and diffuse-porous) wood. Two tasks were considered, including wood patch classification and wood core classification. In these tasks, the proposed model correctly recognized species of almost 93% the wood image patches and 98.7% of wood core images. It also outperformed the state-of-the-art convolutional neural network-based competitor by 9% and 3%, respectively. The influence of the model’s parameters and training set-up on its performance is analyzed in the manuscript to ensure the highest recognition rates of wood species. The source code of the proposed method is released together with the corresponding image dataset to facilitate the reproduction of results.

Evaluation of the Size-of-Source Effect in Thermal Imaging Cameras.

In numerous applications, including current body temperature monitoring in viral pandemic management, thermal imaging cameras are used for quantitative measurements. These require determination of the measurement accuracy (error) and its traceability (measurement uncertainty). Within error estimation, the size-of-source effect (SSE) is an important error source. The SSE is the relation between the physical size of a target and the instrument’s nominal target size. This study presents a direct evaluation of the error due to the SSE. A stable and uniform temperature, generated by blackbodies, was measured by a high-quality thermal imager. To limit the generated radiation, custom-made blocking tiles with different apertures were used. Effects of aperture shapes and positions, camera-target distances and temperature levels on the error were investigated. The study findings suggest that due to the SSE the measured temperatures are too low, especially at longer camera-target distances. The SSE error depends on the number of pixels available and included into the region of interest, for which the accurate measurement is about to be performed. For an accurate temperature measurement, an array of at least 10 × 10 pixels should be exposed to the observed target radiation, while 3 × 3 central pixel area should be included in the temperature calculation.

Pattern of Local Gravitational Force (PLGF): A Novel Local Image Descriptor.

This paper presents a novel local image descriptor called Pattern of Local Gravitational Force (PLGF). It is inspired by Law of Universal Gravitation. PLGF is a hybrid descriptor, which is a combination of two feature components: one is the Pattern of Local Gravitational Force Magnitude (PLGFM), and another is Pattern of Local Gravitational Force Angle (PLGFA). PLGFM encodes the local gravitational force magnitude, and PLGFA encodes the local gravitational force angle that the center pixel exerts on all other pixels within a local neighborhood. We propose a novel noise resistance and the edge-preserving binary pattern called neighbors to center difference binary pattern (NCDBP) for gravitational force magnitude encoding. Finally, the histograms of the two components are concatenated to construct the PLGF descriptor. Experimental results on the existing face recognition databases, texture database, and biomedical image database show that PLGF is an effective image descriptor, and it outperforms other widely used existing descriptors. Even if in complicated variations like noise, and illumination with smaller databases, a combination of PLGF and convolutional neural network (CNN) performs consistently better than other state-of-the-art techniques.