Pattern Recognition and Color Modularity in Mathematics and Art: Knowledge Visualization and Visual Communication

Introduction

Pattern recognition has been with us since the dawn of human consciousness noted M.P. Mattson in Frontiers of Neuroscience: ‘Language-mediated encoding and transfer of auditory and visual patterns is what enabled the rapid evolution of the human brain.’ (2014). In recent history, pattern recognition methodology has become an engine of progress that affects many areas of science, such as computer vision, signal analysis, speech comprehension, and more recently machine learning and artificial intelligence.

Recent progress in the field of Artificial Intelligence and Deep Learning has brought to the forefront the challenge and complexity of our recognition process. Not long ago, Stanford University Computer Science Professor Andrew Ng and Google Fellow Jeff Dean demonstrated a neural network with more than a billion connections as a first step in a larger research project (1991).

When I first studied pattern recognition, most academic studies were focused on numbers, language or symbol sequences, or alternatively, they explored simple to complex black-outline geometric shapes. At the 2012 Bridges Math and Art Conference, Mathematician Reza Sarhangi introduced an intriguing design concept having to do with color modularity. In a lecture centered on the use of mathematics to decode historical Persian tessellations, he brought to our attention the concept of color-contrast modularity as a means to create, enhance, or interpret geometric patterns (2005). Modularity as a technique sorting light, contrast, and gradation of color is also a key element in art and visual communication. It is often used by artists, professionals, and scholars to create, describe, and analyze a design or an artwork. In the 1970s, breaking with the tradition of simply naming colors after a pigment or a subjective description, the brand Liquitex introduced a very popular modular color set based on H. Munsell color system classification of hue, value, and Chroma (1969). This timely expose motivated me to revisit Bongard’s pattern recognition principles in a short experiment based on the technique that Sarhangi described so eloquently in his presentation.

Pattern Recognition And Visual Communication

Pattern recognition is often interpreted as the investigation of structures in a clear geometrical or sequential order (1970) In mathematics, the study of pattern recognition goes from linear algebra and vector space to probability theory and estimation techniques. In the mid-1960s, Russian scientist Mikhail M. Bongard developed a rational approach to solving complex problems of visual recognition and compiled his findings in a book titled Pattern Recognition (1970). The book contains over a hundred line-drawing examples to serve as a practice set for pattern recognition and problem-solving exercises. Accordingly, each problem is composed of two sets of figures positioned side by side, and the viewer is expected to compose a short statement describing the contents of each set separately and with distinct self-standing sentences.