This paper presents the application of the Centroid-Point Distance Histogram, a method for shape representation, in the identification of pollen grains extracted from digital microscopic images. The primary objective is the automated recognition of specific types of pollen, which holds significant importance for individuals suffering from allergies. The proposed method aims to facilitate the development of an intelligent, context-aware system capable of providing real-time information about pollen levels in the air, thereby helping allergy sufferers manage their condition more effectively. The Centroid-Point Distance Histogram combines key features of polar coordinate transformations and histogram analysis, making it particularly suitable for describing shapes such as pollen grains. The methodology was evaluated using real microscopic images and a template matching technique for classification. Initial experimental results, obtained from analyzing 50 air sample images, demonstrated a classification accuracy of 79%. These findings highlight the potential of the method for practical applications in allergen monitoring and pave the way for further research and system optimization.