Senior Product Manager at Upraised | Ex-Citi | IIT Guwahati

productHunt

github

youtube

Sai Bhaskar Devatha • Peerlist Profile

Harmonic and melodic characteristics of a music are the ones that make a sound feel like music. Thus, making them a key distinguisher for an automatic music type classifier like ours. Pitch is a perceptual property of sound that makes humans distinguish between music based on these characteristics. Chroma, a key-component of pitch, is widely used in measuring and analysing these characteristics due to close correlation, while being robust to changes in instrumentation and timbre of the music. Therefore, in this project we extracted and analysed chromagram to study their effectiveness in classifying music into four broad categories namely – Classical, Jazz, Metal and Pop.

Music Genre Classification

Recognised Diabetical retinopathy by reducing the glare and reflections, and aligning the resulting images of the eye using Enhanced Correlation Coefficient Algorithm. Used an order statistical filter for inpainting the resulting images together.

Open Indirect Ophthalmoloscope

This project attempts to adapt the optical camera-oriented CNN to microwave counterpart, i.e, Synthetic aperture radar (SAR). Using Sparse autoencoder and ZCA Whitening, images are pre-processed and later all the processed images after CNN layer are classified using Softmax Regressor.

SAR Target Recognition

this.Survey aims at removing middlemen for aggregating surveys and ensuring peer-to-peer sharing of responses and transacting money through the same. This is achieved by Ethereum Smart Contracts for transaction of money upon meeting the criteria and IPFS - Inter Planetary File System for secure transmission of responses. This system ensures that the current issue of responses, of a survey filler being by other users without his/her permission, is signigicantly reduced.

this.Survey

The Theremin is an electronic musical instrument controlled without physical contact, Our idea is to reconstruct a theremin using Raspberry Pi, Arduinos and sensors like Gyroscope, Accelerometers, Ultrasonic Distance Sensors that generates sine waves of varying frequencies dependent on the movement of hands(which have sensors attached).

Motion based Music Synthesizer

Narrowband, monostatic radar gives information about the objects in space, in the form of a scalar value known as Radar cross-section (RCS). A combination of a CNN and LSTM is used to classify the estimates with a percentage accuracy of 99%.

Monostatic RCS Classification using LSTM and CNN

Project mainly focuses on recreating 'Kogge Stone Adder' in Xilinx using VHDL, used primarily to reduce binary addition of higher bits by creating stepwise carry ahead adder and adding simultaneously to achieve higher speeds.

KSA Implementation using VHDL

Worked with basic principles of Image Processing to analyze images at regular intervals. Analyzed screen colours to identify the laser pointer. These pixel coordinates are sent to the PC which enables the mouse to automatically follow it.