Hybrid Mamdani Fuzzy Rules and Convolutional Neural Networks for Analysis and Identification of Animal Images

Accurate, fast, and automatic detection classification of animal images is challenging, but it much needed for many real-life applications. This paper presents a hybrid model Mamdani Type-2 fuzzy rules convolutional neural networks (CNNs) applied to identify distinguish various animals using different datasets consisting about 27,307 images. The proposed system utilizes detect the image then apply CNN object’s predicate category. was trained tested based on more than 21,846 pictures animals. experiments’ results method offered high speed efficiency, which could be prominent aspect in designing image-processing systems Type 2 characterization identifying fixed moving obtained an accuracy rate recognizing objects 98% mean square error 0.1183464 less other studies. It also achieved very correctly predicting malicious equal recall = 0.98121 precision 1. test’s evaluated F1 Score, percentage 0.99052.