Multi-objective genetic programming for feature learning in face recognition

A Fechner multiscale local descriptor for face recognition

Inspired by Fechner's law, we propose a Fechner multiscale local descriptor (FMLD) for feature extraction and face recognition. Fechner's law is a well-known law in psychology, which states that a human perception is proportional to the logarithm of the intensity of the corresponding significant differences physical quantity. FMLD uses the significant difference between pixels to simulate the pattern perception of human beings to the changes of surroundings. The first round of feature extraction is performed in two local domains of different sizes to capture the structural features of the facial images, resulting in four facial feature images. In the second round of feature extraction, two binary patterns are used to extract local features on the obtained magnitude and direction feature images, and four corresponding feature maps are output. Finally, all feature maps are fused to form an overall histogram feature. Different from the existing descriptors, the FMLD's magnitude and direction features are not isolated. They are derived from the "perceived intensity", thus there is a close relationship between them, which further facilitates the feature representation. In the experiments, we evaluated the performance of FMLD in multiple face databases and compared it with the leading edge approaches. The results show that the proposed FMLD performs well in recognizing images with illumination, pose, expression and occlusion changes. The results also indicate that the feature images produced by FMLD significantly improve the performance of convolutional neural network (CNN), and the combination of FMLD and CNN exhibits better performance than other advanced descriptors.

An Irreversible and Revocable Template Generation Scheme Based on Chaotic System

Face recognition technology has developed rapidly in recent years, and a large number of applications based on face recognition have emerged. Because the template generated by the face recognition system stores the relevant information of facial biometrics, its security is attracting more and more attention. This paper proposes a secure template generation scheme based on a chaotic system. Firstly, the extracted face feature vector is permuted to eliminate the correlation within the vector. Then, the orthogonal matrix is used to transform the vector, and the state value of the vector is changed, while maintaining the original distance between the vectors. Finally, the cosine value of the included angle between the feature vector and different random vectors are calculated and converted into integers to generate the template. The chaotic system is used to drive the template generation process, which not only enhances the diversity of templates, but also has good revocability. In addition, the generated template is irreversible, and even if the template is leaked, it will not disclose the biometric information of users. Experimental results and theoretical analysis on the RaFD and Aberdeen datasets show that the proposed scheme has good verification performance and high security.

Instance Selection-Based Surrogate-Assisted Genetic Programming for Feature Learning in Image Classification

Genetic programming (GP) has been applied to feature learning for image classification and achieved promising results. However, many GP-based feature learning algorithms are computationally expensive due to a large number of expensive fitness evaluations, especially when using a large number of training instances/images. Instance selection aims to select a small subset of training instances, which can reduce the computational cost. Surrogate-assisted evolutionary algorithms often replace expensive fitness evaluations by building surrogate models. This article proposes an instance selection-based surrogate-assisted GP for fast feature learning in image classification. The instance selection method selects multiple small subsets of images from the original training set to form surrogate training sets of different sizes. The proposed approach gradually uses these surrogate training sets to reduce the overall computational cost using a static or dynamic strategy. At each generation, the proposed approach evaluates the entire population on the small surrogate training sets and only evaluates ten current best individuals on the entire training set. The features learned by the proposed approach are fed into linear support vector machines for classification. Extensive experiments show that the proposed approach can not only significantly reduce the computational cost but also improve the generalisation performance over the baseline method, which uses the entire training set for fitness evaluations, on 11 different image datasets. The comparisons with other state-of-the-art GP and non-GP methods further demonstrate the effectiveness of the proposed approach. Further analysis shows that using multiple surrogate training sets in the proposed approach achieves better performance than using a single surrogate training set and using a random instance selection method.

A new artificial intelligent approach to buoy detection for mussel farming

Aquaculture is an important industry in New Zealand (NZ). Mussel farmers often manually check the state of the buoys that are required to support the crop, which is labour-intensive. Artificial intelligence (AI) can provide automatic and intelligent solutions to many problems but has seldom been applied to mussel farming. In this paper, a new AI-based approach is developed to automatically detect buoys from mussel farm images taken from a farm in the South Island of NZ. The overall approach consists of four steps, i.e. data collection and preprocessing, image segmentation, keypoint detection and feature extraction, and classification. A convolutional neural network (CNN) method is applied to perform image segmentation. A new genetic programming (GP) method with a new representation, a new function set and a new terminal set is developed to automatically evolve descriptors for extracting features from keypoints. The new approach is applied to seven subsets and one full dataset containing images of buoys over different backgrounds and compared to three baseline methods. The new approach achieves better performance than the compared methods. Further analysis of the parameters and the evolved solutions provides more insights into the performance of the new approach to buoy detection.