LEDTD: Local edge direction and texture descriptor for face recognition☆
Introduction
Face recognition, as one of the most focused research topic in image processing, pattern recognition and computer vision, has been widely applied in many fields, such as information security, smart cards, entertainment, law enforcement, video surveillance and human–computer interaction. Image feature extraction serves as one of the most critical steps for face recognition. Although numerous approaches have been proposed and tremendous progress has been made, during the past decades, it is still could not perform as well as desired under uncontrolled conditions. Therefore, how to extract discriminative and robust features is of vital importance to face recognition.
Generally, the two-dimensional image feature extraction methods in image representation could be broadly summarized into two categories based on their properties, i.e., holistic methods and local methods. The holistic methods generally extract features from a facial image by treating the image as a whole. Principal component analysis (PCA) [1], linear discrimination analysis (LDA) [2], independent component analysis (ICA) [3], locality preserving projection (LPP) [4], local linear embedding (LLE) [5], local discriminant embedding (LDE) [6], marginal Fisher analysis (MFA) [7], discriminant simplex analysis (DSA) [8], nonnegative graph embedding (NGE) [9], clustering-guided sparse structural learning (CGSSL) [10] and robust structured subspace learning (RSSL) [11] are the typical ones of this kind. These methods are liable to be influenced by face image pose, illumination, scale and so on, and variations in these factors can largely degrade its recognition performance. The local methods usually consider several regions or sets of isolated points, from which features for classification are extracted. Classical methods such as local binary pattern (LBP) [12], [13], scale-invariant feature transform (SIFT) [14], [15], speeded-up robust features (SURF) [16], weber local descriptor (WLD) [17], Weber local binary pattern (WLBP) [18], monogenic binary coding (MBC) [19], histograms of local dominant orientation(HLDO) [20], enhanced local directional pattern (ELDP) [21], farthest point distance (FPD) descriptor [22], rotation-invariant fast feature (RIFF) [23], edge orientation difference histogram (EODH) [24] have been widely examined. Compared with holistic methods, local methods are distinctive and invariant to many kinds of geometric and photometric transformations, and have been gaining more and more attention because of their promising performance.
Being one of the representative local image descriptors, local binary pattern (LBP) was first introduced by Ojala et al. [12], and it has shown a high discriminative ability for texture classification due to its invariance to monotonic gray level changes. Afterwards, many variants of LBP have been introduced to further improve its performance. However, the feature of all these methods being coded into the bit-string is prone to change due to noise or other variations.
Considering that Kirsch compass mask enhances the useful information like edge texture and meanwhile suppresses the external noise effect, recently it has been widely used for image feature extraction. Jabid et al. [25] proposed local directional patterns (LDP), which is an eight-bit binary code calculated by first comparing the absolute edge response values derived from different directional Kirsch masks. Then the top k prominent values are selected and the corresponding directional bits are set to 1, the remaining bits are set to 0. Finally, convert the binary number into a decimal one, and the decimal value is the corresponding image pixel LDP expression. Zhong and Zhang [26] proposed the enhanced local directional patterns (ELDP), which improved the LDP in the following two aspects. First, take the sign of the Kirsch edge response into consideration, which means two opposite trends (ascending or descending) of the gradient and contain some more discriminant information. Second, only the most and the second most prominent edge response directions are take into the local pattern coding. Kang et al. [27] proposed the structured local binary kirsch pattern (SLBKP), which quantify the eight edge responses into two four-bit binary codes according to the predefined threshold. Rojas Castillo et al. [28] proposed local sign directional pattern (LSDP), similar to the ELDP, the only difference is that it codes the most and the least prominent edge response directions. Rivera et al. [29], [30] proposed local directional texture pattern (LDTP), which is the mixture coding of direction number of local most prominent Kirsch mask edge response and intensity differences along two greatest edge responses directions.
In this paper, we propose a novel discriminative and robust image descriptor, local edge direction and texture descriptor (LEDTD), for face recognition. The main novelty of our descriptor can be summarized as follows. (1) Compared with image gray-scale value, the edge direction is more stable than intensity, and the use of edge direction feature makes our descriptor more robust against illumination variations and noise by operating in the gradient domain. (2) Edge responses are not equally important for image feature extraction. We choose directions of the maximum and minimum response, which explicit the gradient direction of bright and dark areas in the neighborhood, to represent local image pixel edge information. (3) Apart from the directional features, local XOR operator is applied to encode image edge direction texture features, which convey power image discriminative information. (4) Our LEDTD makes full use of the center pixel edge direction and surrounding eight neighbor pixels texture information while existing image descriptor LDTP only utilizes four neighbor pixels to encode image local structure. Therefore, LEDTD retains more local structure information than the LDTP. Experimental results demonstrate the superiority of our LEDTD compared with the state of the art image representation approaches.
The remainder of the paper is organized as follows. Section 2 presents our LEDTD based image representation in detail and a brief review of SVM based face image classification method. Section 3 conducts experiments to evaluate the performance of the proposed method. Section 4 concludes the paper.
Section snippets
Local edge direction and texture descriptor based image representation
Our descriptor fully capture the local image edge direction and texture information for feature coding. The overall framework of the proposed approach is illustrated in Fig. 1.
Databases and experimental setup
The AR face image database [33] contains more than 4000 face images of 126 subjects (70 men and 56 women) with different facial expressions, illumination conditions, and occlusions. For each subject, 26 images were taken in two separate sessions (two weeks interval between the two sessions). A subset that contains 100 subjects (50 male and 50 female) is chosen in our experiments and the original images are normalized to pixels.
The Extended Yale B face database [34] consists of 2414
Conclusion
In this paper, we have proposed a simple and easy to compute image descriptor, local edge direction and texture descriptor (LEDTD), for face image representation. The main findings of the work are as follows: (1) The proposed LEDTD exploits both image edge direction and texture information available locally, which is evidenced by the improved performance. (2) The WPCA method can further improve the recognition performance of the proposed image descriptor. The experiments have been conducted on
References (36)
- et al.
WlbpWeber local binary pattern for local image description
Neurocomputing
(2013) - et al.
Discriminative histograms of local dominant orientation (d-hldo) for biometric image feature extraction
Pattern Recognit.
(2013) - et al.
Face recognition with enhanced local directional patterns
Neurocomputing
(2013) - et al.
Farthest point distancea new shape signature for fourier descriptors
Signal Process.: Image Commun.
(2009) - et al.
Rotation-invariant fast features for large-scale recognition and real-time tracking
Signal Process.: Image Commun.
(2013) - et al.
color-siftapplication to image retrieval based on codebook
Signal Process.: Image Commun.
(2014) - et al.
Face recognition with enhanced local directional patterns
Neurocomputing
(2013) - et al.
Eigenfaces for recognition
J. Cognitive Neurosci.
(1991) - et al.
Eigenfaces vs. Fisherfacesrecognition using class specific linear projection
IEEE Trans. Pattern Anal. Mach. Intell.
(1997) - et al.
Face recognition by independent component analysis
IEEE Trans. Neural Netw.
(2002)
Regularized locality preserving projections and its extensions for face recognition
IEEE Trans. Syst. Man Cybern. Part B: Cybern.
Discriminant locally linear embedding with high-order tensor data
IEEE Trans. Syst. Man Cybern. Part B: Cybern.
Graph embedding and extensions: a general framework for dimensionality reduction
IEEE Trans. Pattern Anal. Mach. Intell.
Classification and feature extraction by simplexization
IEEE Trans. Inf. Forensics Secur.
Clustering-guided sparse structural learning for unsupervised feature selection
IEEE Trans. Knowl. Data Eng.
Robust structured subspace learning for data representation
IEEE Trans. Pattern Anal. Mach. Intell.
Cited by (18)
TPCNN: Two-path convolutional neural network for tumor and liver segmentation in CT images using a novel encoding approach
2021, Expert Systems with ApplicationsA novel pipeline framework for multi oriented scene text image detection and recognition
2021, Expert Systems with ApplicationsCitation Excerpt :Additionally, the edge direction is defined by the mask that generates the maximum magnitude. A new method is introduced in this study that can yield better accuracy in character recognition based on previous studies (Ranjbarzadeh & Saadi, 2020; Ryu, Rivera, Kim, & Chae, 2017; Rivera et al., 2015; Rose et al., 2015; Li et al., 2016; Luo et al., 2016; Liu et al., 2016; Karimi, Kondrood, & Alizadeh, 2017). Local word directional pattern (LWDP) is an efficient texture analyzer algorithm that computes the gradient of the adjacent pixels in eight main orientations inside a patch (a 5 × 5 window).
LNPSS: SAR image despeckling based on local and non-local features using patch shape selection and edges linking
2020, Measurement: Journal of the International Measurement ConfederationCitation Excerpt :Each pixel of images all utilizes these aforementioned eight masking templates to perform the convolution [49]. The mask which efficiently produces the maximum value is properly reflected in the edge direction [50–53]. By choosing adaptive thresholding of magnitude, the only noticeable and long edges have remained leading to better segmentation of regions.
Automated liver and tumor segmentation based on concave and convex points using fuzzy c-means and mean shift clustering
2020, Measurement: Journal of the International Measurement ConfederationCitation Excerpt :The edge scale corresponds to the maximum value obtained from the convolution of each mask with the image. The maximum value calculated by one of the eight masks defines the edge direction [24–27]. This part is useful for calculating the concave and convex points and adjusting the contrast using the mean-shift algorithm.
Illumination-insensitive image representation via synergistic weighted center-surround receptive field model and weber law
2017, Pattern RecognitionCitation Excerpt :However, without fully utilizing some useful local region direction information and adjacent pixels’ importance information related to their central pixel, all of these above methods have suffered from limited performance improvement. Therefore, some improved methods are further proposed, including local derivative pattern (LDP) [44], local Gabor XOR patterns (LGXP) [45], local orientation adaptive descriptor (LOAD) [46], dominant LBP (DLBP) [47], complete ELDP (CELDP) [48], eight local directional patterns (ELDP) [49], local directional number pattern (LDN) [50], adaptive homomorphic eight local directional patterns (AH-ELDP) [51], local edge direction and texture descriptor (LEDTD) [52], logarithmic fractal dimension (LFD) [53], pyramid local binary pattern (PLBP) [54] and patterns of oriented edge magnitudes (POEM) [55]. Beyond that, scholars also present some local descriptors based on Gabor filters due to its robustness against local distortions caused by illumination variance.
Pair of projections based on sparse consistence with applications to efficient face recognition
2017, Signal Processing: Image CommunicationCitation Excerpt :As one of the traditional hot issue in computer vision, face recognition (FR) identifies the face image without label information according to the faces stored with labels automatically. There have been many works [1–5] on FR recently. Normally, the common approaches to FR are confronted with the efficiency problem due to the high-dimension of the face images.
- ☆
Fully documented templates are available in the elsarticle package on CTAN.