Acne is a common skin condition caused by dead skin cells and excess oil blocking the hair follicles. It affects people of all ages and is frequently caused by hormonal fluctuations. Acne can appear in many forms, such as whiteheads, papules, pustules, cysts, nodules, blackheads, or pimples. Traditionally, acne detection has been visually performed with the help of a dermatologist. However, recent advances in communication and technology have emerged a new direction for acne detection tasks. The research interest in automated acne-type detection is increasing rapidly due to the advantages of cost-effectiveness and time efficiency of computer-aided diagnosis (CAD). Various computer-aided approaches using machine learning and deep learning algorithms have recently been proposed to automate acne-type detection from dermatological images. Despite the advancement, some challenges still exist such as color shifting caused by inconsistent illumination, scale variations, high-density distribution, detecting acne from images of different body parts, and identifying the percentage disbursement of different acne types in human skin. We have developed a recognition system to detect acne type and overcome the challenges accurately. To achieve this, we have used YOLOv8, a computer vision model specifically designed to identify the acne mentioned above types precisely. We have conducted extensive experiments on datasets from DermNet NZ and Kaggle as well as we have collected our dataset. The model’s average accuracy is almost 53%. The evaluation results indicate that our suggested method surpasses other methods for identifying different types of acne.