Software Systems Research Lab (SSRL) offers software development and research services in diverse domains including web applications, enterprise systems development, prototyping, and smart-phone applications. SSRL is currently expanding its services by engaging in research and development activities in the areas of smart-phone applications and games. Open-source technologies are extensively utilized in software development in order to achieve cost and time efficiencies. SSRL also offers training in web application development and mobile app and game development.
Following are some of the software systems and training programs developed by SSRL:
- UHands: Emergency Management System built on location-based technologies.
- Wi-CAM: TR-069 and OMA-DM Auto Configuration Server for remote provisioning, configuration, and firmware upgrade for Consumer Premise Equipments.
- UAMS: University Academics Management System
- Certificate Program in Mobile App and Game Development.
- Library management system using RFID technologies.
- Bio-metric attendance management system
- Configuration & customization of Open ERP.1
Data Science Lab focuses to cater to the challenges of current age which may be termed as the era of “data big bang”. Driven by the internet economy, mobile phone, cheaper hardware and the Internet of Things (IoT), the user and other sensory devices are continuously generating a lot of data. As data size increases, the demand for multi-scale approaches in transforming data to knowledge also becomes very important.
Research in Data Science involves the design of intelligent algorithms and the development of decision-making models to form risk management systems and process modeling systems. Interpreting data and visualizing it to define patterns and extract knowledge can help businesses to compete with other competitors. The lab focuses on both structured and unstructured data analytics for clustering, classification, and association rule mining to identify trends and make useful predictions.
The lab is currently focused on developing a knowledge extraction framework that can be applied to multiple types of textual data including news articles, scientific literature, social media, and police investigation reports. The lab aims to bring together researchers, industry experts, and students to provide a platform for flourishing the field of data science in Pakistan.
- Finding new approaches for data collection, integration, and data/information sharing technologies
- Develop new statistical and mathematical algorithms, prediction method, modeling methods, compaction schemes.
- Seeking a new way to derive useful, reliable, and verifiable information from big and complex data sets, by using advances in information processing, integration, machine learning, data mining, compression, and visualization of data.
- Focus on data science research and education that address the challenges of large data sets, high consumption rates, short analysis time windows, different content and media types, and contradicting, incorrect, and missing information.
- Develop scalable data processing systems, and showcase their solutions to challenging real-world use cases of relevance to science, industry, and society.
Automated Ground Vehicle Research Lab (AGVRL) focuses on the research and development of automation of mobile platforms for surveillance systems. There are two main goals of this lab, the first one is to design the localization, path planning and navigation algorithms for the mobile platform. And the second is to deploy the latest state of the art deep learning models like suspicious activity detection, harmful object detection, and person identification, etc.
The group has made available dome public domain biological resources such as Genbank through the intranet.
- Development of an extensible data model
- Development of the prototype-based software development methodology for bioinformatics
- Proposing metrics to evaluate Data Provenance
- Methodology for bioinformatics
- Proposing metrics to evaluate Data Provenance.
Computer Vision and machine learning lab focuses on the research and development of image processing tools, computer vision algorithms, natural language processing and applications helpful to the community from the field of computer vision.
Dedicated research staff comprising 2 PHDs and over 10 full-time research officers and undergraduate/postgraduate interns, the lab is involved in the development of core APIs required for the surveillance systems including face recognition, emotions, and action recognition and scene settings.
The purpose of this research lab is to delve deeply inactive area of research of computer vision and machine learning and provide the solutions of local problems to our community. Moreover, it is a wonderful opportunity for young aspirants to get a working knowledge of computer vision systems under the supervision of experts. CVML Lab offers various solutions to public places like metro stops, railway stations, bus stops, airports, parks and roads, forbidden zones like offices and railway tracks, suspicious zones and generation of video data to a textual format. Some of the ongoing projects are summarized below.
In the paradigm of surveillance systems, the lab has recently started a project of “Automatic Surveillance of Video Streams” which is also funded by the National ICT R&D Fund and also has collaboration with foreign institutes and local organizations. The project is focused on the recognition of suspicious activities from the public places and forbidden zones by extracting the high level spatial and temporal features from video streams. The project also aims to explore the limitations of image processing and computer vision algorithms in different environments. The development of this project will be demonstrated through surveillance applications.
Another paradigm of Natural Language Generation (NLG) and Natural Language Processing is focused on the reduction of space required by video streams generated by surveillance cameras by converting multimedia into textual format. In short, Computer vision and machine learning lab are capable to help the various areas of life which are not limited to provided domains.