Fast and robust rotation invariant object detection with Joint Color Channel and Hierarchical Binary Pattern

In this paper, we propose a method for fast and robust rotation invariant object detection using Joint Color Channel (JCC) and Hierarchical Binary Pattern (HBP) to be used as the classifier in well-known cascade AdaBoost. The cascade structure is efficiently designed according to the attribute of the features for fast object detection. To evaluate our proposed method, we use a drum dataset collected in the real industrial environment. Drums have a variety of colors and textures depending on their type and have pose variations when they are tilted by humans. The experimental results on the real images show the applicability and high efficiency of the proposed method.