Rade L. Profile

EXPLORE

0.0

(0 reviews)

You're now following

Error following user.

This user does not allow users to follow them.

You are already following this user.

Your membership plan only allows 0 follows. Upgrade here.

Successfully unfollowed

Error unfollowing user.

You have successfully recommended

Error recommending user.

Something went wrong. Please refresh the page and try again.

Email successfully verified.

$30 USD / hour

beograd,

serbia

$30 USD / hour

It's currently 10:22 PM here

Joined January 3, 2024

0 Recommendations

Rade L.

@lazarevicr94

0.0

(0 reviews)

0.0

$30 USD / hour

beograd,

serbia

$30 USD / hour

N/A

Jobs Completed

N/A

On Budget

N/A

On Time

N/A

Repeat Hire Rate

SENIOR SOFTWARE ENGINEER

- Highly skilled and motivated AI Developer with over 10 years of experience in machine learning, deep learning, natural language processing (NLP) and data analysis, seeking to leverage my skills and experience to help the company drive innovation and improve business outcomes. - Proficient in machine learning, and Deep learning. - Building an effective conversational AI system by combining expertise in NLP, ML, and Generative AI. - Analytical and process-oriented data scientist, with in-depth knowledge of database types, research methodologies, and big data capture, curation, manipulation, and visualization. - Passionate about research and innovation in computer vision.

See all PHP Developers in Serbia

Contact Rade L. about your job

Portfolio

TeleHealth/Telemedicine Services.
We Provide “Telemedicine Software Development”, “HealthCare App Development”, “Streaming Telehealth Technology”, “Telemedicine Integrations”, “Store and Forward Telemedicine Services”, “HealthCare Software Solutions”

Tele-Health Solution Design and Development

License Plate Detector

Drowsiness Detection System with audio alert system is developed to detect eye closure detection, yawning and head movement in parallel by applying transfer learning on pre-trained VGG16 Model.
The real-time data of these detection is then posted on IoT platform for analysis.

IoT Enabled Drowsiness Detection and Alert System

Grad-CAM Bounding-Boxes with GRAD-CAM Grad-CAM:

Visual Explanations from Deep Networks via Gradient-based Localization Creation of Bounding Boxes (Rectangle and Polygon) using a Grad-CAM Heatmap

CAM: Learning Deep Features for Discriminative Localization

Visual Explanation from Deep Neural Networks

The project's objective is that:

Given a phone book page, Output a CSV file with the OCR result for all columns.

The challenge is that the page has a very dense structure and the rows and text are so close to each other so straight OCR techniques will give poor results and we won't be able to distinguish the column number or the row number of the OCR'ed text.

So, we have created two computer vision algorithms:
1) Segments all the columns on the page to know the boundaries of every column.
2) Segments all the rows of each column.

In this way, we can know the exact location (#column, #row) of the OCR'ed text.

Finally, for every row crop it and pass it to the DL OCR model “Easy OCR” to get the OCR text then save the results to the output CSV file.

Phone Book OCR

Real-time analytics for traffic from CCTV footage.

The system is able to process a number of cameras with accurate stats.

System is made using Machine Learning and Deep Learning Algorithms with PyTorch in Python.

Features include:
- Vehicle Classification
- Direction Classification
- Classified Directional Counting
- Speed
- Traffic Density

This project is currently in deployment stage with government organizations from Turkey and Australia.

Vehicle Detection

Tele-Health Solution Design and Development

License Plate Detector

IoT Enabled Drowsiness Detection and Alert System

Visual Explanation from Deep Neural Networks

Phone Book OCR

Vehicle Detection

Reviews

Changes saved

No reviews to see here!

Experience

Sunflower Lab Remote

Jun 2018 - Nov 2023 (5 years, 5 months)

Accomplishments: - Developed recommendations for a leading online retailer with over 1 million products. Used machine learning algorithms like Matrix Factorization and KNN to analyze past purchase histories and generate personalized recommendations displayed on the product detail page and homepage. - Developed a virtual assistant for an online educational platform to answer student questions, provide homework help and schedule reminders for assessments.

AGENCY NOVELUS

May 2014 - Apr 2020 (5 years, 11 months)

- Customized License Plate Detector is designed by applying Transfer Learning on Yolov3 weights using coco datasets. - Built perception systems using cameras, lidar and radar to detect pedestrians, vehicles and Traffic signs for a self-driving car prototype. - Implemented project is to construct a streaming data pipeline by making use of the real-time financial market data API provided by Finnhub.