BHARATH C H

I hold a B.Tech in Electrical and Electronics Engineering, specializing in VLSI design, signals and systems, and machine learning. My projects include designing a Finfet-based op-amp and developing a facial recognition system using convolutional neural networks. I have expertise in MATLAB, demonstrated through EEG signal processing and image processing applications. Internships at IISc and Bosch have equipped me with skills in integrating machine learning with hardware systems, preparing me to tackle complex challenges in VLSI and intelligent systems.

• Led a team and won the first prize in the Bosch-IEEE EV 2.0 hackathon. Designed an electric vehicle and implemented advanced reinforcement learning techniques. This resulted in the successful development of a state-of-the-art motor control system that demonstrated the immense potential of AI-driven technologies in the field of sustainable transportation.
• Took part in a community-driven effort to clean and rejuvenate the Uttarahalli Lake in Bengaluru, demonstrating a commitment to environmental conservation and the well-being of local ecosystems.
• Was selected from taluk level and participated in Bengaluru-South district level Competition in chess, Badminton, and kabaddi, Bengaluru, September 2019

• Patent in AI, for an image processing system using neural networks, enabling real-time object recognition and tracking with greater accuracy and speed. (Is submitted and is under review)
• Data Acquisition and Pre-Processing of EEG Signals using Brainwave Sensor, 3rd International Conference on Signal and Data Processing (ICSDP), 09/2023
• Fire Fighting Robot, International Journal of All Research Education & Scientific Methods (IJARESM), 05/2023
• Power System Protection, International Journal of Innovative Science and Research Technology (IJISRT), 05/2023

• Door Unlock Using Facial Recognition and Fingerprint Scanner, Role: Hardware Integration

This was part of an internship in the ISFCR club of PES University where we designed a door unlock using Raspberry Pi, a solenoid lock, a fingerprint scanner, and a camera associated with Raspberry Pi and detected faces using a convolutional neural network. This can be incorporated in every household application and also offices. 

• Text Extraction and Translation of Sign Boards, Role: Python Developer

 It is difficult for people to translate sign boards from different regions that are written in their native language. Hence we developed a text extraction and translation of sign boards and this was done using various image processing techniques namely MSER and OCR in Matlab. To make sure the OCR extracts texts in proper shape we adopted a few pre-processing using aspect ratio, eccentricity, Euler number, extent, and solidity. Matlab and Jupyter Notebook (python) were used. 

• Design and analysis of Finfet-based op-amp, Role:Circuit designer

 Designed a stable 9 Transistor Op-Amp configuration with a high gain. Optimized the amplifier to have better stability, lesser noise, and lesser power dissipation for the same gain. Cadence Virtuoso was utilized extensively and 9T configuration was found to produce lesser noise and required configuration. This op-amp can be used for any low-power application after fabrication. 

• Data Acquisition and Pre-processing of EEG Signals for Detection of Autism and Degree of Depression Using Neural Network, Role: Software developer

 As part of my final year project, I developed a portable, low-cost EEG signal acquisition device. We applied signal processing techniques, including adaptive filtering and wavelet transform, to reduce noise and analyze signal frequency bands in Matlab. We used this device to gauge depression levels based on signal characteristics, such as centroid distances. Additionally, we employed neural networks and empirical mode decomposition to assess autism, training and validating the model with SVM and k-fold cross-validation.