Harini Bachu

Hi, I'm B.HARINI,as a driven and curious undergraduate student majoring in Computer Science(AI & ML), I am passionate about exploring the latest advancements in technology and applying them to real-world problems. With a strong foundation in programming languages such as Python and Java, I am eager to continue learning and growing as a developer. I am excited to contribute to innovative projects and collaborate with like-minded individuals who share my enthusiasm for technology. I’m passionate about how machines learn, think, and solve real-world problems. I'm currently focusing on building my fundamentals in programming, data science, and AI.

🛠 Tech Skills Existing Skills:

Python, C/C++, Java (Basics)

Planning to Learn:

TensorFlow, Keras, NLP (spaCy, NLTK)

Cloud Platforms (AWS, GCP), MLOps basics

Web Deployment (Flask, Streamlit), Docker

💼 Projects I Worked On

Pet face classification: Project Description: Pet Face Classification

Objective:

The objective of this project is to develop a machine learning model that can accurately classify pet faces into different breeds or categories. This project aims to explore the application of computer vision and deep learning techniques in pet face recognition.

Dataset:

The project will utilize a dataset of images of pet faces, including dogs, cats, and potentially other animals. The dataset will be sourced from publicly available repositories or collected through web scraping. Each image will be labeled with the corresponding breed or category.

Methodology:

1. Data Preprocessing: Images will be resized, normalized, and potentially augmented to increase the size and diversity of the dataset.
2. Feature Extraction: Deep learning techniques, such as Convolutional Neural Networks (CNNs), will be used to extract features from the images.
3. Model Training: A classification model will be trained using the extracted features and labels.
4. Model Evaluation: The performance of the model will be evaluated using metrics such as accuracy, precision, recall, and F1-score.

Potential Applications:

1. Pet Identification: The model can be used in pet identification systems, allowing pet owners to identify their pets or find lost pets.
2. Pet Breed Classification: The model can be used in pet breed classification systems, helping breeders, shelters, or rescue organizations identify the breed of a pet.
3. Animal Behavior Analysis: The model can be used as a starting point for analyzing animal behavior, such as detecting emotions or facial expressions.

Technical Requirements:

1. Programming Language: Python will be used as the primary programming language.
2. Deep Learning Framework: A deep learning framework such as TensorFlow or PyTorch will be used to build and train the model.
3. Computer Vision Library: A computer vision library such as OpenCV will be used for image processing and feature extraction.

Expected Outcomes:

1. Accurate Classification: The model is expected to accurately classify pet faces into different breeds or categories.
2. Robustness: The model is expected to be robust to variations in image quality, lighting, and pose.
3. Generalizability: The model is expected to generalize well to new, unseen data.

Landmark detection:

Project Description: Landmark Detection

Objective:

The objective of this project is to develop a computer vision system that can accurately detect and identify landmarks in images or videos. Landmarks can include features such as facial landmarks (e.g., eyes, nose, mouth), object landmarks (e.g., corners, edges), or scene landmarks (e.g., buildings, monuments).

Dataset:

The project will utilize a dataset of images or videos with annotated landmarks. The dataset can be sourced from publicly available repositories, such as the 300W facial landmark dataset or the Oxford Buildings dataset.

Methodology:

1. Data Preprocessing: Images or videos will be preprocessed to enhance quality, remove noise, and normalize features.
2. Landmark Detection: A deep learning-based approach, such as Convolutional Neural Networks (CNNs) or Recurrent Neural Networks (RNNs), will be used to detect landmarks.
3. Model Training: The model will be trained using the annotated dataset, with a suitable loss function and optimization algorithm.
4. Model Evaluation: The performance of the model will be evaluated using metrics such as accuracy, precision, recall, and Intersection over Union (IoU).

Potential Applications:

1. Facial Recognition: Landmark detection can be used in facial recognition systems, enabling applications such as security surveillance, identity verification, and social media tagging.
2. Object Recognition: Landmark detection can be used in object recognition systems, enabling applications such as robotics, autonomous vehicles, and industrial inspection.
3. Augmented Reality: Landmark detection can be used in augmented reality applications, enabling interactive and immersive experiences.

Technical Requirements:

1. Programming Language: Python will be used as the primary programming language.
2. Deep Learning Framework: A deep learning framework such as TensorFlow, PyTorch, or Keras will be used to build and train the model.
3. Computer Vision Library: A computer vision library such as OpenCV will be used for image and video processing.

Expected Outcomes:

1. Accurate Landmark Detection: The model is expected to accurately detect and identify landmarks in images or videos.
2. Robustness: The model is expected to be robust to variations in lighting, pose, and occlusion.

📜 Certifications

Python for Data Science (IBM – Coursera)

1stop.ai

TCS-ION CAREER EDGE Certification

GUVI Certification (guvi.in)

🌟 Aspirations I aspire to become a skilled AI/ML Engineer and work on impactful projects. I also aim to gain expertise in Cloud Computing to deploy AI solutions at scale. Long-term, I’d love to explore research in deep learning and contribute to open source. .