Face recognition model for age invariant

الملخص EN

In this paper we design and implement a new model to recognize images for human face for age invariant by using the Scale Invariant Feature Transform (SIFT) algorithm.

The proposed model transform an image into a large collection of local feature vectors, each of which is invariant to image and then using the Structural Similarity Index Measure (SSIM) algorithm after adjusts the light effect using the Retinex algorithm.

The images used from the public data base FG-NET.

These images are divided into three groups depending on the age gap between two images for the same person (two images per person in each group).

The results obtained from this system were encouraging compared with the results obtained via other studies using the same data base and different strategies.

Available database represent a great challenge for the face recognition across age progression.

Also the performance of age invariant face recognition system is dropped as the age gap increases.

Future work is described that outlines using another database contains more than two images for each subject , and collecting the images in one group with increasing age gap.