Abstract
There are a wide variety of types of garbage, and combustible garbage discharged from ordinary households includes various types of garbage such as kitchen waste, paper, plastics, and plants, each of which has a different moisture content and calorific value. In order to stabilize the combustion of the refuse incineration furnace, it is necessary to stir the refuse well in the refuse pit before throwing it in to keep the water content and calorific value constant. In recent years, the spread of image processing technology and advances in AI technology have led to the development of advanced automatic stirring technology that matches the state of the waste in the pit1). The key technology is to grasp the mixed state of dust from the pit image, and the statistical variation of the pixel color is used2). However, since there is no decisive index that expresses the mixed state of garbage, this research developed an index of "uniformity" of garbage using the concept of relative entropy used in information theory. Then, we applied it to the pit image and confirmed its effectiveness as an index for uniform agitation.
