Car image recognition with lacked input data

A car type recognition system, using methods of the R&FHPR/FF (Rough and Fuzzy Hierarchical Pattern Recognition using Fixation Feedback), infer just about correctly based on precise input data. Although, it is difficult, features are extracted completely in general using image processing methods. For systems using fuzzy inference, if input datum is lacking, the system does not work well. A framework of a modified fuzzy inference method with lacking input data is introduced. Experimental results are provided. In the proposed method which is modified from A. Mamdani´s (1975) fuzzy inference, the result of each rule is the adjustment for the purpose of protection from the influence of lacking input data. The adjustment of the resulting fuzzy labels at each rule uses the degree of importance which is set up in the rules manually. In the experimental results, the system can infer well with lacking input data using this method. The experiments used a set of eight fuzzy rules (five input and eight output), and when all combinations of input data are lacking, the system infers so correctly