A Gaussian Filter-Based Feature Learning Approach Using Genetic Programming to Image Classification

Abstract

Image classification is an important task in computer vision. Image features are often needed for classification, but the majority of feature extraction methods require domain experts and human intervention. To learn image features automatically from the problems being tackled is more effective. However, it is very difficult due to image variations and the high dimensionality of image data. This paper proposes a new feature learning approach based on Gaussian filters and genetic programming (GauGP) for image classification. Genetic programming (GP) is a well-known evolutionary learning technique and has been applied to many visual tasks, showing good learning ability and interpretability. In the proposed GauGP method, a new program structure, a new function set and a new terminal set are developed, which allow it to detect small regions from the input image and to learn discriminative features using Gaussian filters for image classification. The performance of GauGP is examined on six different data sets of varying difficulty and compared with four GP-based methods, eight traditional approaches and convolutional neural networks. The experimental results show GauGP achieves significantly better or similar performance in most cases. Further analysis on example GP programs demonstrates the good interpretability of GauGP and the importance of Gaussian filters in GauGP in terms of feature learning and image classification.

Genetic Programming With Image-Related Operators and a Flexible Program Structure for Feature Learning in Image Classification

“…These methods use the normalised raw pixel values of images as inputs to train the classifiers. The three SVM methods are LBP+SVM, HOG+SVM and SIFT+SVM [11], which use LBP, HOG, or SIFT features as inputs of SVMs for classification. The LBP, HOG and SIFT features are extracted by the methods described in Table II.…”

“…When applied to new tasks with unknown types of images, these methods may not be effective [2]. Therefore, feature learning methods have been proposed to automatically learn features from images for classification [2,11]. With a learning process, the features can be optimised or fine-tuned to achieve good performance.…”

“…With a learning process, the features can be optimised or fine-tuned to achieve good performance. e.g., the maximum classification accuracy on the training set [11]. Compared with the hand-crafted features, the learned features are often more effective for a wide range of image classification [2,12].…”

An automated ensemble learning framework using genetic programming for image classification

“…In recent years, many image-related operators such as Gaussian filter, mean filter, Sobel filter, and HOG have been developed as functions in GP to extract/learn high-level features for effective classification [3,17,27]. These image-related operators were very helpful for learning important features [3,17,27]. Based on the framework of GP for feature learning in [3,27], this paper uses a set of image-related operators for feature learning with the consideration of efficiency and effectiveness.…”

“…These image-related operators were very helpful for learning important features [3,17,27]. Based on the framework of GP for feature learning in [3,27], this paper uses a set of image-related operators for feature learning with the consideration of efficiency and effectiveness.…”