Genetic Programming With a New Representation to Automatically Learn Features and Evolve Ensembles for Image Classification

Bi, Ying; Xue, Bing; Zhang, Mengjie

doi:10.26686/wgtn.13158311.v1

Cited by 10 publications

(17 citation statements)

References 11 publications

(16 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This transfer learning approach has the ability to solve complex tasks that most other algorithms remain unable to tackle, as shown by the results. Most recently, Bi et al [12] proposed a GP method to learn novel features automatically, and simultaneously evolve an ensemble for image classification. This method uses commonly used classification algorithms and image-related operators such as Gabor filter, Laplacian filter, LBP, and HoG, to evolve ensembles of classifiers for classification.…”

Section: Measure Formulamentioning

confidence: 99%

“…Two feature manipulation techniques, feature selection and feature construction, can be employed in such cases which help to pick important features and construct new informative features provided the original set of features, respectively, to improve classification accuracy [9]- [11]. Feature extraction, on the other hand, transforms the original set of features into a reduced representation set [12]. Most recently, convolutional neural networks (CNNs) have gained immense popularity in dermoscopy image analysis.…”

Section: Introductionmentioning

confidence: 99%

“…In addition to classification, GP has been explored extensively for feature selection and construction [9]- [11]. In image analysis, a wide range of applications have utilized GP, including object detection [19], feature extraction [12], feature construction [11], and classification [20], [21]. Recently, a GP method [11] has been developed using texture features for skin cancer detection from dermoscopy images.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automatically Diagnosing Suspicious Regions in Skin Images Using Two-Stage Genetic Programming

Ain¹,

Al-Sahaf²,

Xue³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Developing a computer-aided diagnostic system for detecting various skin malignancies from images has attracted many researchers. Unlike many machine learning approaches such as Artificial Neural Networks, Genetic Programming (GP) automatically evolves models with flexible representation. GP successfully provides effective solutions using its intrinsic ability to select prominent features (i.e. feature selection) and build new features (i.e. feature construction). Existing approaches have utilized GP to construct new features from the complete set of original features and the set of operators. However, the complete set of features may contain redundant or irrelevant features that do not provide useful information for classification. This study aims to develop a two-stage GP method where the first stage selects prominent features, and the second stage constructs new features from these selected features and operators such as multiplication in a wrapper approach to improve the classification performance. To include local, global, texture, color, and multi-scale image properties of skin images, GP selects and constructs features extracted from local binary patterns and pyramid-structured wavelet decomposition. The accuracy of this GP method is assessed using two real-world skin image datasets captured from the standard camera and specialized instruments, and compared with commonly used classification algorithms, three state-of-the-art, and an existing embedded GP method. The results reveal that this new approach of feature selection and feature construction effectively helps improve the performance of the machine learning classification algorithms. Unlike other black-box models, the evolved models by GP are interpretable, therefore the proposed method can assist dermatologists to identify prominent features, which has been shown by further analysis on the evolved models.

show abstract

Section: Measure Formulamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatically Diagnosing Suspicious Regions in Skin Images Using Two-Stage Genetic Programming

Ain¹,

Al-Sahaf²,

Xue³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Evolutionary algorithms are known for their powerful global search ability and nondifferentiable requirement [7]. They have been applied to many tasks in machine learning, such as optimizing the architectures of DNNs [8][9][10] and finding the optimal model of ensemble [11][12][13]. However, no evolutionary algorithms have been developed to optimize the structure of deep forest.…”

Section: Related Workmentioning

confidence: 99%

Evolving Deep Forest with Automatic Feature Extraction for Image Classification Using Genetic Programming

Xue

Zhang

2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Deep forest is an alternative to deep neural networks to use multiple layers of random forests without back-propagation for solving various problems. In this study, we propose a genetic programming-based approach to automatically and simultaneously evolving effective structures of deep forest connections and extracting informative features for image classification. First, in the new approach we define two types of modules: forest modules and feature extraction modules. Second, an encoding strategy is developed to integrate forest modules and feature extraction modules into a tree and the search strategy is introduced to search for the best solution. With these designs, the proposed approach can automatically extract image features and find forests with effective structures simultaneously for image classification. The parameters in the forest can be dynamically determined during the learning process of the new approach. The results show that the new approach can achieve better performance on the datasets having a small number of training instances and competitive performance on the datasets having a large number of training instances. The analysis of evolved solutions shows that the proposed approach uses a smaller number of random forests over the deep forest method.

show abstract

“…However, this method has shown inferior classification results on large image classification datasets. Most recently, Bi et al [42] proposed a GP method to learn novel features automatically and simultaneously evolve an ensemble for image classification. This method uses commonly used classification algorithms and image-related operators such as Gabor filter, laplacian filter, LBP, and HOG, to evolve ensembles of classifiers for classification.…”

Section: Related Work To Image Classification Using Gpmentioning

confidence: 99%

Genetic Programming based Feature Manipulation for Skin Cancer Image Classification

Ain¹

View full text Add to dashboard Cite

<p>Skin image classification involves the development of computational methods for solving problems such as cancer detection in lesion images, and their use for biomedical research and clinical care. Such methods aim at extracting relevant information or knowledge from skin images that can significantly assist in the early detection of disease. Skin images are enormous, and come with various artifacts that hinder effective feature extraction leading to inaccurate classification. Feature selection and feature construction can significantly reduce the amount of data while improving classification performance by selecting prominent features and constructing high-level features. Existing approaches mostly rely on expert intervention and follow multiple stages for pre-processing, feature extraction, and classification, which decreases the reliability, and increases the computational complexity. Since good generalization accuracy is not always the primary objective, clinicians are also interested in analyzing specific features such as pigment network, streaks, and blobs responsible for developing the disease; interpretable methods are favored. In Evolutionary Computation, Genetic Programming (GP) can automatically evolve an interpretable model and address the curse of dimensionality (through feature selection and construction). GP has been successfully applied to many areas, but its potential for feature selection, feature construction, and classification in skin images has not been thoroughly investigated. The overall goal of this thesis is to develop a new GP approach to skin image classification by utilizing GP to evolve programs that are capable of automatically selecting prominent image features, constructing new high level features, interpreting useful image features which can help dermatologist to diagnose a type of cancer, and are robust to processing skin images captured from specialized instruments and standard cameras. This thesis focuses on utilizing a wide range of texture, color, frequency-based, local, and global image properties at the terminal nodes of GP to classify skin cancer images from multiple modalities effectively. This thesis develops new two-stage GP methods using embedded and wrapper feature selection and construction approaches to automatically generating a feature vector of selected and constructed features for classification. The results show that wrapper approach outperforms the embedded approach, the existing baseline GP and other machine learning methods, but the embedded approach is faster than the wrapper approach. This thesis develops a multi-tree GP based embedded feature selection approach for melanoma detection using domain specific and domain independent features. It explores suitable crossover and mutation operators to evolve GP classifiers effectively and further extends this approach using a weighted fitness function. The results show that these multi-tree approaches outperformed single tree GP and other classification methods. They identify that a specific feature extraction method extracts most suitable features for particular images taken from a specific optical instrument. This thesis develops the first GP method utilizing frequency-based wavelet features, where the wrapper based feature selection and construction methods automatically evolve useful constructed features to improve the classification performance. The results show the evidence of successful feature construction by significantly outperforming existing GP approaches, state-of-the-art CNN, and other classification methods. This thesis develops a GP approach to multiple feature construction for ensemble learning in classification. The results show that the ensemble method outperformed existing GP approaches, state-of-the-art skin image classification, and commonly used ensemble methods. Further analysis of the evolved constructed features identified important image features that can potentially help the dermatologist identify further medical procedures in real-world situations.</p>

show abstract

Genetic Programming With a New Representation to Automatically Learn Features and Evolve Ensembles for Image Classification

Cited by 10 publications

References 11 publications

Automatically Diagnosing Suspicious Regions in Skin Images Using Two-Stage Genetic Programming

Automatically Diagnosing Suspicious Regions in Skin Images Using Two-Stage Genetic Programming

Evolving Deep Forest with Automatic Feature Extraction for Image Classification Using Genetic Programming

Genetic Programming based Feature Manipulation for Skin Cancer Image Classification

Contact Info

Product

Resources

About