Image Clarity Evaluation Method Considering Visual Characteristics

Abstract

　The surface appearance of products affects buying intention. In particular, image clarity is an important appearance quality factor that affects the elegance of the products, for example, automobile and smartphone. The correlation between the conventional evaluation methods and visual inspection is poor. In addition, these methods damage the surface of products due to a contact measurement. The purpose of paper is to develop a new image clarity evaluation method that is high correlation with visual inspection and a non-contact measurement for solid and metallic color samples. To develop the new image clarity evaluation method, first, the measurement device composed of a spectral camera and a lighting device for projecting an edge pattern on an object was constructed. Next, subjective experiments on image clarity were conducted using Sceffeʼs paired comparison. The new evaluation model based on the spatial frequency characteristic of the projected pattern image and human visual characteristic was proposed. Furthermore, the surface color was added to the model. As a result, the contribution ratio of the new evaluation model and the subjective score was very high.