A general vehicle logo system consists of two major parts, i.e.
the logo detection module and the logo recognition module.
In this paper, the main focus is the classification and recognition of vehicle logo.
The proposed method uses moment invariants for feature extraction of vehicle logo images.
Two types of moment invariants namely Tchebichef and Legendre moment invariants are compared and analyzed
in terms of their recognition accuracy.
The coarse segmented vehicle logo images are obtained from a public dataset of [3] while the fine logo images are manually segmented from the original dataset.
Features from both input vehicle logo image types are extracted by moment invariants and subsequently
classified using Minimum-Mean Distance (MMD).
The outline of the proposed system is shown in Figure 1.