Dr. Surbhi Sharma

Associate Consultant

Gurugram, Haryana, India17 yrs experience

Highly StableAI Enabled

Key Highlights

9+ years of experience in Machine Learning and Computer Vision.
Led data science initiatives at FiftyFive Technologies.
Published multiple research papers in SCI journals.

Stackforce AI infers this person is a Data Science and Computer Vision expert with a strong research background.

Contact

s.surbhi1986@gmail.com LinkedIn

Skills

Core Skills

Data ScienceMachine LearningComputer Vision

Other Skills

CCustomer Relationship Management (CRM)CybersecurityDigital Image ProcessingFurnitureGenerative AIIBM WatsonImage ProcessingLaTeXMatlabMicrosoft OfficeNatural Language Processing (NLP)OpenCVPowerPointProgramming

About

Experienced Ph.D. Professional in Computer Vision and Machine Learning with multiple Publications in SCI journals. After Ph.D, experienced Data scientist with 9+ years experience in research and developing ML models according to client's requirements and available resources. Delivered projects like AI-based coaching system, AI-based restaurant management system, PCB identification, Automatic physical damage detector for mobile phones, background removal, object detection, and tracking. Proficient in Python, TensorFlow, Keras, Sql, Big Query, Computer Vision, Deep Learning, Predictive Analysis and Modelling Looker, Flask, Docker, Power BI, and MLOps. Adept in learning new tools

Experience

Ibm

Managing Consultant, AI

Jul 2024 – Present · 1 yr 8 mos · India · Hybrid

At IBM, I'll be working alongside a team of brilliant minds, leveraging cutting-edge technologies specially in DTT, and driving innovative solutions to help our clients navigate the ever-evolving landscape. Our mission is to empower businesses to reach their full potential by delivering tailored, strategic insights and transformative results using AI.

Fiftyfive technologies

2 roles

Head of Data Science

Promoted

Apr 2023 – Jun 2024 · 1 yr 2 mos

Led strategic vision and execution for data science initiatives, fostering innovation (RL, Gen AI, MLops) and collaboration in a high-performing team.
Managed end-to-end data science projects, optimizing workflows for timely delivery and alignment with business goals.
Facilitated cross-functional collaboration, translating complex technical concepts for effective communication with non-technical stakeholders.
Mentored, and retained data science talent.
Drove innovation through R&D, staying abreast of emerging technologies to enhance predictive analytics and machine learning models.

Data ScienceMachine Learning

Senior Data Scientist

Nov 2020 – Apr 2023 · 2 yrs 5 mos

Cashify

Data Scientist

Dec 2019 – Nov 2020 · 11 mos · Gurgaon, Haryana

National institute of technology kurukshetra

2 roles

Ph.D. Research Scholar

Aug 2013 – Jul 2019 · 5 yrs 11 mos · Kurukshetra

Projects:
1. Forensic Analysis of Images – Collected, studied and interpreted large image datasets. Performed analysis to find problems in existing models. Developed new models to increase accuracy & robustness in classifying Computer Generated Images , Tampered Images & Natural Images.
 Spliced Image Classiﬁcation & Spliced Region Localization – Designed & implemented new model to classify SIs from NIs using SVM classifier. LDP based Feature extraction and selection used to localize spliced regions with accuracy > 98.55 %. Model is also robust against post splicing operations (Gaussian blur, AWGN, JPEG compression & Scaling).
Published in International Journal of Image, Graphics and Signal Processing.
 Region Duplication Detection – Improved detection accuracy rate (to > 0.99) in case of basic manipulations & decreased false positive rate (to < 0.01) using proposed Dominating Direction based features. The proposed model used Contourlet Transform (CT), SVD, KNN, Morphological Operation & worked well in presence of Gaussian blur, AWGN, JPEG compression, Rotation & Scaling with high accuracy. Project can be scaled up for automated detection of tampering in images & video.
Published in The Imaging Science Journal.
 Digital Image Forensics (DIF) – Developed model to classify given image datasets into CGIs, TIs & NIs along with localization of tampered regions. Classified CGIs & NIs using periodicity in image datasets & ELM classifier. Periodicity checked using LOG high pass filter on green channel of RGB images, worked as features for classification & achieved Accuracy > 97.07%. CFA & sensor noise based features extracted from NIs to localize spliced region & achieved Precision, Recall Values of 98.51% & 95.15% respectively. Detection of Region duplication done using GLCM features, KNN, M-SAC & morphological opening.
Published in Optik – Journal (Elsevier).
2. Vehicle Number Plate Recognition
3. Object Recognition
4. Wireless Gesture Control Robot

Digital Image ProcessingComputer VisionMachine Learning

Assistant Professor

Aug 2008 – Jun 2013 · 4 yrs 10 mos · Kurukshetra, India

1. Implemented Projects using Computer vision , Machine Learning, Neural Networks and Fuzzy Systems.
2. Taught subjects like Image Processing, Artificial Intelligence, Machine Learning, Computer Vision, Neural Networks, Fuzzy Systems to B.Tech. and M. Tech students.
Project
1. Content based Image Retrieval – Features extracted using color moments, auto correlogram and Gabor Wavelet.