COMPARATIVE ANALYSIS OF OBJECT CLASSIFICATION ALGORITHMS: TRADITIONAL IMAGE PROCESSING VERSUS ARTIFICIAL INTELLIGENCE – BASED APPROACH

Abstract

In the current era of advanced digital technologies, form recognition is integrated into numerous applications, from computer vision to industrial automation. This paper focuses on a comparative analysis of two distinct form recognition algorithms, namely harnessing the power of artificial intelligence (AI) and image processing techniques. The research is motivated by the need to address the trade-off between speed and complexity in form recognition, with a center on real-world applicability. Traditional image processing-based form recognition approaches often require complex coding, substantial domain expertise, and significant computational resources. This complexity can hinder rapid adaptation to changing requirements and the addition of new forms. The aim is to explore whether AI-powered algorithms can offer a more efficient and versatile alternative, reducing the barriers to entry for form recognition tasks. The primary goal of the paper is to compare the performance of AI-based form recognition with image processing-based methods in terms of speed and accuracy. The second goal is to assess the ease of adapting AI-based algorithms to new forms without extensive code changes. Two form recognition algorithms were designed and implemented, one based on artificial intelligence and a second relying on image processing. The AI-powered algorithm uses neural network architecture trained on a predefined dataset of forms. The image processing algorithm employs edge detection and contour analysis techniques.