Beauty business is a huge business in most of the countries. That is because people care about their look, their face and body more than ever before. However, acne has been the problem for facial treatment for decades. The first step in facial treatment is the acne counting and quantification. The dermatologist count the number of acne before and after the treatment to see whether the patients respond well to the treatment or not. The counting is the vital process in the acne treatment. However, dermatologist normally count the number of acne manually and mark them on a facial shape on paper (as shown in Fig.1). There were many publication proposed to tackle this problem.
All the automatic processes used the process and the basic steps from image processing and detection system[1-3]. In this publication, it is involved with face detection[4], blob detection[5], and color segmentation[6]. We develop a program to automatically count the number of acne which is also useful to help in the medical treatment. It will count the number of detected acne that helps the doctor see the development of acne of each patient or the response to the treatment. The doctor can also compared the weekly result very clearly that acne on the patient's face is getting better or worse. So, the doctor can perform analysis and select appropriate approach to treatment.
Figure1. Images of doctor write down the disease information of patients by hand written.
The developing algorithms of acne detection worked on the collection of data taken locally (as shown in Fig 2). Those pictures were taken into Photoshop by expert and then the experts mark them on a different Photoshop layer (Fig 3) and we used that layer (Fig 4) as our gold standard or ground-truth that we use to compare our result to. Only for cheek area that is under consideration. The ground truth images are shown in Fig 4.