Current deep convolution neural network (CNN) has shown to achieve superior performance on a number of computer vision tasks such as image recognition, classification and object detection. The deep network was also tested for view-invariance, robustness and illumination invariance. However, the CNN architecture has thus far only been tested on non-uniform illumination invariant. Can CNN perform equally well for very underexposed or overexposed images or known as uniform illumination invariant? This is the gap that we are addressing in this paper. In our work, we collected ear images under different uniform illumination conditions with lumens or lux values ranging from 2 lux to 10,700 lux. A total of 1,100 left and right ear images from 55 subjects are captured under natural illumination conditions. As CNN requires considerably large amount of data, the ear images are further rotated at every 5o angles to generate 25,300 images. For each subject, 50 images are used as validation/testing dataset, while the remaining images are used as training datasets. Our proposed CNN model is then trained from scratch and validation and testing results showed recognition accuracy of 97%. The results showed that 100% accuracy is achieved for images with lumens ranging above 30 but having problem with lumens less than 10 lux

No abstract

No abstract

No abstract

Smartphones are constantly being upgraded with new technologies and applications. Since the implementation of touch screen technology, users have had an easier time navigating through the functions of their mobile devices. Although the latest technology has brought us improvements, a handful of users have been experiencing difficulty in using the touch screen devices. A visually impaired person uses android phones are facing difficulties with touch screen technology. Many researches have taken place to solve the stated problem and the optimal solution is the creation of applications that functions fully with voice and minimizing touch based selections. This project is based on improving the accessibility features in android smartphones by implementing the use of text to speech as well as speech to text functions for visually impaired. To achieve this objective an application is developed using android studio which uses speech input to traverse the application, also to read incoming calls and messages. The developed system is tested with an evaluation test to public. Based on result it is analysed overall satisfaction in using the application is high, with few improvement in features such as rate of speech and call & message problem. In future work, the system can be developed to run in all platforms, adding more features to daily life activities such as calendar event handling, and integration of system into reading calls and messages from other apps such as WhatsApp and Facebook messenger.