Acne Detector

People now a days facing acne diseases and some time they did not concerned to thier doctors resulting major diseases in future. We create an web application through which user can registered yourself as a patient and doctors can registered as a doctors.    We create a portal on whi

Project Title

Acne Detector

Project Area of Specialization

Artificial Intelligence

Project Summary

We create a portal on which doctors shares thier article for thier researches as a Blog and patient can book thier OPD slots by paying some amount.

Now come to the main point that the purpose of application that we collect initial data of acne and thier treatment as well and save them in database. User insert an image and application gives the name of medicine. The whole process is done by past experience on the basis of learning.

Project Objectives

We are facing acne diseases day by day. Our application will helps patients to identify the type of acne and treatment as well. It will also communication medium between doctors and patient virtually.

Project Implementation Method

Implementation of user infer face (UI/UX)

Registered in application like sign/up

Doctors registered in application like sign/up

User have option to give payment method

Portal on which doctors shares thier article on research basis

Collect initial data of the type of acne

Collect medicine data for treatment

User insert image of acne and application suggested the medicine by past experience

Book appointment slots for doctor if user want

Benefits of the Project

People can easily identify thier type of disease by clicking an image without any time consume and cost.

Technical Details of Final Deliverable

Lesion counting is assessed on both the face (from the forehead, left and right cheeks and chin above the jaw line, excluding the nose) and the trunk (shoulders, upper back and upper anterior chest) looking for the number of non-inflammatory lesions (open comedones and closed comedones), inflammatory lesions (papules, pustules) and other lesions (nodules and cysts). Other skin conditions that can mimic acne include bacterial folliculitis, miliaria, perioral dermatitis, pesudofolliculitis barbae, rosacea and seborrheic dermatitis. These entities may have in common the presence of pustules or papules, however, the absence of comedones rule out acne in differential diagnosis

In this paper we focus on image processing on acne that is how we identify acne on faces using mobile application that send the data through medium (client/server or tcp/ip) to the web that integrated MATLAB.

Identification and classification of acne lesion;

The RGB image is converted to CIE L*a*b* color space where luminance can be separated from color values. The a* channel represents the redness of the pixel color independent of its luminance and can be used to robustly detect the increase of skin redness and identify the location of the lesions. A new image is generated using a lowpass 2D Gaussian filter that blurs the image and removes detail and noise. The Gaussian kernel standard deviation is defined to ensure that lesions are Once the lesions have been separated from other features on the face, acne lesions can be classified as either papules and pustules. The classification is performed using the differences between papules and pustules. Pustules can be differentiated from the papule by their whitish and liquid center. During the acne identification process, the center of pustule region stays at zero in the binary image. The white center has a relatively lower redness difference compared to the surrounding region with high redness due to the skin inflammation. This results in a binary mask with a circular shaped hole inside of the pustules (white area). The Euler number for the binary image is used to automatically separate the binary regions with and without a hole (papules and pustules). The Euler number is the total number of objects in the binary region minus the total number of holes in that region. An Euler number equal to one represents a papule and an Euler number equal to zero represents a pustule

The performance of the identification of lesions is reported by finding the number of true positives (number of lesions identified correctly), number of false positives (number of lesions identified incorrectly), and number of false negatives (number of missed lesions). The performance of classification is calculated by determining the confusion matrix for classification and finding the accuracy using

1. ???????????????????????????????? = (???????????????? ???????????????????????????????? (????????) +???????????????? ???????????????????????????????? (????????) )/????????????????????