International Journal of Computer Science and Engineering

1-8

Authors: Apurva Pavaskar, Saurabh Dawkhar, Diksha Kamble, Praneet Mugdiya & S. F. Sayyad

Commercially viable photography began in the late 1820s, but the effective on-set of color photography started as late as 1970s. During this period of over a century, photographs captured were mostly black and white. Colorization plays a vital role in representing the true virtue of real-world manifestations. The human eye perceives color and often remembers information about an object based on its coloration. Colorization problem is difficult to solve without manual adjustment. The proposed systems develop colored versions of gray scale images that closely resemble the real-world versions. The use of Artificial Neural Networks in the form of Convolutional Neural Networks (CNN) and Generative Adversarial Networks (GAN) to learn about features and characteristics through training allows for assigning plausible color schemes without human intervention.

15-18

Authors: Supriya H S, Sumanth Alva R, Suprith K P, Vikas Kumar L & P Kyshan Neheeth

Clinical decision support systems are computer based automated systems developed with the aid of AI and ML for supporting and improving the accuracy of clinical decision-making processes. These systems are used by clinicians in making diagnostic decisions and treatment plans. It is able to simulate expertise and express logical reasoning for making assertions. Many inexperienced clinicians are not well confident in certain autistic cases because their observed diagnosis and the calculated grade may not be always similar. Availability of diagnostic experts to provide clinical expertise is also a problem in the diagnosis of autism children. Hence, there is a need of a computer assist system comprised of experience and skill of a clinician, which can advance the power of existing diagnostic method. The computer-assisted system will help to confirm the assessment decisions of clinicians. Research works about the application of machine learning techniques for the development of autism assessment and grading in this area is very much required. This work proposes an approach which uses machine learning techniques for autism grading.