Local-Gravity-Face (LG-face) for Illumination-Invariant and Heterogeneous Face Recognition

ABSTRACT:

This paper proposes a novel method called local-gravity-face (LG-face) for illumination-invariant and heterogeneous face recognition (HFR).LG-face employs a concept called the local gravitational force angle (LGFA). The LGFA is the direction of the gravitational force that the center pixel exerts on the other pixels within a local neighborhood. A theoretical analysis shows that the LGFA is an illumination-invariant feature, considering only the reflectance part of the local texture effect of the neighboring pixels. It also preserves edge information. Rank 1 recognition rates of 97.78% on the CMU-PIE database and 97.31% on the Extended Yale B database are achieved under varying illumination, demonstrating that LG-face is an effective method of illumination-invariant face recognition. For HFR, when faces appear in different modalities, LG-face produces a common feature representation. Rank 1 recognition rates of 99.96% on the CUFS database, 98.67% on the CUFSF database, and 99.78% on the CASIA-HFB database show that the LG-face is also an effective method for HFR. The proposed method also performs consistently in the presence of complicated variations and noise.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

• For good face recognition performance, the variations among faces within each class (intra-class variations) should be small and those between different classes (inter-class variations) should be large. Different illumination conditions pose a vital hindrance to face recognition because they can dramatically affect the appearance of facial image, hence strongly increasing the inter-class variation. It has been proven that changes in the illumination of the face images significantly degrade the performance of a face recognition system.

• The existing methods generally fall into three broad categories. Preprocessing, Modeling and Invariant feature representation

DISADVANTAGES OF EXISTING SYSTEM:

• Existing system does not handle the illumination variations.

• These methods do not yield fully satisfactory results.

• In some cases, preprocessing removes too much useful information, and as a result, the performance in the subsequent recognition phase is degraded.

• The disadvantages of these model-based approaches include the need for multiple face images under different illumination conditions and 3-D shape information during training.

PROPOSED SYSTEM:

In this paper, we propose an illumination-invariant feature extraction method based on the direction of the local gravitational force. We call this directional feature the Local Gravitational Force Angle (LGFA), and the method is called Local-Gravity-Face (LG-face). We also prove that the LG-face method is illumination-invariant and that its recognition performances on existing challenging databases are quite impressive. The main features of the proposed LG-face approach are as follows:

• The features are illumination-invariant.

• There is no need to estimate the illumination components (L); they are automatically discarded.

• No smoothing preprocessing of the images is performed, and therefore, there is no loss of texture, i.e., the reflection components (R) and edges are preserved.

• Because it is based on the local gravitational force effect, the LGFA is considered a local feature.

• The relative positions of key facial features are not modified.

• The method reduces the intra-class variations among faces in the same class and enhances the inter-class variations among faces in different classes.

ADVANTAGES OF PROPOSED SYSTEM:

• LG-face can easily be used as a preprocessing tool to eliminate variations in illumination, shadows, and variations in facial expression and to enhance facial edge features. Therefore, it has an extensive range of potential applications, from coping with illumination variations to heterogeneous face recognition. It may also have promising prospects in other fields of image processing, a possibility that is worth further investigation.

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

• System            : Pentium Dual Core.

• Hard Disk       : 120 GB.

• Monitor          : 15’’ LED

• Input Devices : Keyboard, Mouse

• Ram               : 1GB.

SOFTWARE REQUIREMENTS:

• Operating system : Windows 7.

• Coding Language : MATLAB

• Tool                      : MATLAB R2013A

REFERENCE:

Hiranmoy Roy and Debotosh Bhattacharjee, Senior Member, IEEE, “Local-Gravity-Face (LG-face) for Illumination-Invariant and Heterogeneous Face Recognition”, IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, VOL. 11, NO. 7, JULY 2016.