Image Analytics


Over 80% of data in enterprises comprise of unstructured datasets in form of text, images and videos. These datasets are very information rich but lot tougher to mine and analyze compared to structured datasets. Today’s advances in cloud computing and deep learning have made it possible to extract information out from such datasets to large extent, opening up opportunities to bring the benefits this vision to our daily lives. It is possible to analyze moods of different people in single photographs, identify security threats by analyzing images and video feeds, detect leakages or machine failures looking at images, recognize handwritten content in scanned forms, and recognize license plate and vehicle occupancy detection systems. These applications are already deployed and generate value for enterprises.

Valiance has been working with enterprises in mining image datasets using deep learning. Our team runs algorithms on cloud with GPU machines for training deep learning models. Together with our machine learning and engineering expertise we can help you mine image datasets.

Content Digitization

Automatically extracting data fields from scanned application forms using image processing & deep learning./span>

Facial Recognition

Identify individuals using our proprietary image recognition algorithms. Algorithms can be trained for variety of business scenarios.

Case Studies

Handwritten Text Recognition

Valiance has created information extraction platform using Matlab and Tensorflow (deep learning platform from Google) with Mysql as backend data store. .

Image Summarization

Valiance has created a platform using Python and Microsoft Cognitive Toolkit CNTK(deep learning framework by Microsoft Research) to summarize the images.