| Skin is the largest organ of the human body.As a barrier between the human body and the outside world,it plays a vital role in life activities.The traditional manual method of facial skin diagnosis is not only time-consuming and labor-intensive,but also has high requirements on the technology and experience of the diagnostic doctor due to the complexity of facial skin appearance,and is highly subjective.The use of special spectroscopy and computer image processing technology to quantitatively and automatically evaluate the human facial skin,not only can reduce labor costs,but also improve the diagnostic accuracy by analyzing the invisible information to the naked eyes,and can objectively analyze the facial skin condition.Both disease prevention and clinical diagnosis have broad application prospects.Human facial skin contains four features,namely appearance features,surface features,dermis features and invisible features.According to the physicochemical properties of these features and physical characteristics of different spectra,four special lights are used(standard light,parallel polarized light,vertical polarized light and Wood light)for image acquisition of human facial skin images.Then it uses digital image processing technology to identify and extracts these features,analyzes the color,position and shape of features,calculates nine quantitative indicators of skin features.Based on this principle,a human face skin detection system was developed and implemented to help doctors diagnose facial skin quality and disease.Through experimental verification,the experimental results show that the four special spectra used in this system have obvious enhancement effects on different skin feature images,and can give quantitative evaluation information of human facial skin,which can meet the requirements of practical applications. |
