Color correction in photos taken by smartphones

• Main Author: Lai, Liangfa
• Other Authors: Deepu Rajan
• Format: Final Year Project
• Language: English
• Published: 2013
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
• Online Access: http://hdl.handle.net/10356/51967
• Institution: Nanyang Technological University

Mobile phones are now an essential technology for people of all ages which brings along the trend of taking and sharing photos. These mobile phone cameras come with some white balance settings but these settings can be inaccurate due to variance of the lightings in different scenarios. Such inaccuracy may affect what the user wants to share through the photo. Therefore the user will have to process the image to produce an image more realistically coloured. However, not all mobile phone users possess such photo editing skills. The purpose of this project is to write an Android application that takes the hassle of editing photo away from the user. The user will only need to decide which photo he deems as having realistic colours and select it as the sample photo. Subsequently the application will allow the user to take photos and process them to match the settings in the chosen sample photo. One of the approaches used in the application for editing the photo is based on the colour of the subject that appears in both the sample photo and the photo to be edited. Face recognition tools will also be used to detect the face of the subject in both photos and retrieve crucial information from the nose pixel. This information of the nose pixels will be calculated and adjusted to match each other so that the colour of the result will be closer to desirable. Another approach used in the application allows the user to tap on a pixel which is supposed to be white and it will readjust the all the pixels based on the selected pixel. We have done literature review on various methods to edit the colours of the photo and designed an algorithm suitable to be used in the mobile application. The algorithm uses minimal resources which allow it to execute effectively on mobile devices. With the algorithm, we developed the mobile application and tested it on Android Ice Cream Sandwich and Android Jellybean devices. The application is written to ease the user in editing photos in this fast paced world by offering them two modes of photo editing, using sample photo or tapping white pixel. As future improvements, we would recommend that the interface to be more user-friendly and users should be able to achieve satisfactory result within just a few taps.