Current activities:

 Medical Image Processing

Quality improvements of the echo-cardio graphic imaging with the help of a developed suite of software tools is under active field trial in the hospitals. The cost of equivalent commercial medical imaging systems with such sophisticated imaging and signal processing capabilities are rather steep. Moreover, in India, many hospitals, health centers are still using comparatively older versions of such echocardiography machines. Under these activities, the sparkled screen images from these existing echocardiograph machines will be improved in clarity and contrast in a PC. 

Apart from these kinds of desirable noise cleaning aspects, better clinical feature extraction like Motion-mode (M-Mode) display, extraction of Left Ventricle (LV) Boundary , ejection fraction (E.F)  calculation, contrast enhancement of the images are the important attributes of the proposed project . For static organs, femur length (FL) and bi-parietal diameter (BPD) measurements of fetus in the womb is at advanced stage of development on demand from a leading National Medical Instrument Manufacturer. 

 Face Recognition

 The feature-based Face Recognition Software under development is expected to stretch to the utmost state of art technologies in the field of image processing and pattern recognition. A major responsibility will also be to develop a facial image database consisting of many ethnic Indian faces for authentication, verification and benchmarking any system and research component in face recognition area.

At present the software tool is under development to identify a human being from its frontal facial image. The main feature of the software is the following:

·          Normalization of facial input facial image in terms of contrast and size.

·          Preprocessing and noise cleaning.

·          Robust feature extraction.

·          Matching with a database.

Past Activities

Small Target Tracking by Image Processing Techniques

Automatic target recognition (ATR) generally refers to the autonomous or aided target detection and recognition by computer processing of data from a variety of sensors. In this particular application the optical system receives IR data, which is then converted into an image by an array of detectors. This image contains the signature of the target contaminated by clutter and noise. The objective of the project is 

• Detection of Target at low SNR environment.

• Tracking the Target.

• Generation of error signals to guide to control unit.

The development of algorithm and testing on simulated scene is complete.

Testing of algorithms on real data is in progress.

Assessment of Future Trends in Ultra-sonography and a Globally Competitive Design of USG Machine – a roadmap for India:.

In this proposed pilot study, an extensive survey will be made regarding the possibility to come up with a globally competitive design on ultra-sonography (USG) machine.

The study will incorporate following search areas.

• Technology involved in this activity.

• Future trends in this area of research.

• Finding out suitable public/private sectors (who are already engaged in USG development/service), who can implement the developed technology in the form of a product.

• Marketability of the developed product.

• Identification of the important milestones based on which the work of development will proceed.