Abstract
In the field of computer science, data mining is a hot topic. It is a mathematical method for identifying patterns in enormous amounts of data. Image mining is an important data mining technique involving a variety of fields. In image mining, art image organization is an interesting research field worthy of attention. The classification of art images into several predetermined sets is referred to as art image categorization. Image preprocessing, feature extraction, object identification, object categorization, object segmentation, object classification, and a variety of other approaches are all part of it. The purpose of this paper is to suggest an improved boosting algorithm that employs a specific method of traditional and simple, yet weak classifiers to create a complex, accurate, and strong classifier image as well as a realistic image. This paper investigated the characteristics of cartoon images, realistic images, painting images, and photo images, created color variance histogram features, and used them for classification. To execute classification experiments, this paper uses an image database of 10471 images, which are randomly distributed into two portions that are used as training data and test data, respectively. The training dataset contains 6971 images, while the test dataset contains 3478 images. The investigational results show that the planned algorithm has a classification accuracy of approximately 97%. The method proposed in this paper can be used as the basis of automatic large-scale image classification and has strong practicability.
Highlights
Art image classification is one of the most significant research topics with great significance in which a great deal of information is available to people with the help of advancements in computer technology [1, 2], multimedia technology [3], and network technology [4]
Given the possibility of overfitting due to the lack of data, a stochastic gradient descent (SGD) algorithm is used to enhance the objective purpose while avoiding local optimization [25]. e classifier model used in this paper is depicted in Figure 1, which shows how an support vector machine (SVM) classifier has been used to assign test images
To assess the efficacy and standardization of image classification, high-quality, standard, open, and universal image datasets are used. ese datasets consist of 10471 images to conduct classification experiments, and these images are randomly categorized into 2 parts, which are used as training data and test data, respectively. e training data consist of 6971 images, while the test data consist of 3478 images
Summary
In the field of computer science, data mining is a hot topic. It is a mathematical method for identifying patterns in enormous amounts of data. Image mining is an important data mining technique involving a variety of fields. Feature extraction, object identification, object categorization, object segmentation, object classification, and a variety of other approaches are all part of it. E purpose of this paper is to suggest an improved boosting algorithm that employs a specific method of traditional and simple, yet weak classifiers to create a complex, accurate, and strong classifier image as well as a realistic image. This paper uses an image database of 10471 images, which are randomly distributed into two portions that are used as training data and test data, respectively. This paper uses an image database of 10471 images, which are randomly distributed into two portions that are used as training data and test data, respectively. e training dataset contains 6971 images, while the test dataset contains 3478 images. e investigational results show that the planned algorithm has a classification accuracy of approximately 97%. e method proposed in this paper can be used as the basis of automatic large-scale image classification and has strong practicability
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
