Inverter air conditioner intelligent system design

Abstract

As the computers are becoming a part of people's daily lives, human-computer interaction has started to gain more importance. Even though the traditional human computer interfaces like the keyboards and mouses are used frequently, they are making the information and command flow limited. Recently, gestures are not only gaining a more important role in the environment-human relations but also in human-machine interactions as well. The studies on the meaning of the gestures and what kind of information they carry are increasing their places in researches on human-machine interface communication. The aim in this study is to provide an easy and user independent human -machine interface system in air conditioners. In the proposed system, in order to recognize hand gestures, first image preprocessing techniques and noise filters have been applied to images taken from camera. These processes provide clarification of image. After preprocessing, Histogram of Oriented Gradients (HOG) algorithm is used along with the skin color detection made via image processing techniques applied in order to the recognition of hand gestures. After then, the hand gestures have made meaningful as the obtained images are given as inputs to artificial neuron network and an algorithm, which ensures fan and temperature settings with 92% success level in inverter air conditioners, is developed