semiconductor

Deep Neural Network Model Development and Training

Client Overview

US based leading semiconductor manufacturer offering microcontrollers and processors for sensors, analog ICs and connectivity.

Business Need

  • The client was looking for a technology partner to help them in proving the application of newly developed chipsets for Machine Learning at edge to be launched in the market

VOLANSYS Contribution

  • Designed the handwritten digit recognition application using powerful supervised deep learning technique – Convolution Neural Network (CNN)
  • Developed Caffe model using AlexNet architecture and trained using MNIST database of 50,000 images
  • Developed script in Python to export model parameter & converted using CMSIS-NN library to deploy on edge – i.MXRT1062
  • Developed application in C on NXP i.MXRT1062 for capturing the image, pre-processing and executing deep learning model for digit classification
  • Analyzed the performance of the model by tweaking different parameters and layers of the AlexNet architecture
  • Performed testing of the model with 10,000+ images

Solution Diagram

EXT_229

Benefits Delivered

  • Accelerated client’s product launch timeline by 30% with hands-on experience in machine learning domain
Similar Success Stories