The visibility of images displayed on mobile devices can vary significantly according to the lighting conditions. Generally, the brightness of a mobile device is adjusted to enhance visibility according to the intensity of illumination. However, different brightness settings can be more suitable to different types of content even under the same illumination. Accordingly, this paper presents a method to adjust a device's brightness according to the features of the displayed app images. We started by performing two subjective tests under various lighting conditions for selecting the features concerning visibility for several apps screenshots and for selecting a satisfactory range of device brightness for each of those. Then, we analyzed the relationship between the app image features and the satisfactory brightness levels. Subsequently, the images are categorized by using two features: the average brightness and the distribution ratio of advancing colors related to satisfactory brightness. The optimal device brightness for each category is then selected by the maximum frequency of satisfactory device brightness. Experimental results show that the categorized app images with optimal device brightness have high satisfaction ratio under various light conditions.