SPRADC9 july   2023 AM62A3 , AM62A7

 

  1.   1
  2.   Abstract
  3.   Trademarks
  4. 1Introduction
    1. 1.1 Defect Detection Demo Summary
    2. 1.2 AM62A Processor
    3. 1.3 Defect Detection Systems
    4. 1.4 Conventional Machine Vision vs Deep Learning
  5. 2Data Set Preparation
    1. 2.1 Test Samples
    2. 2.2 Data Collection
    3. 2.3 Data Annotation
    4. 2.4 Data Augmentation
  6. 3Model Selection and Training
    1. 3.1 Model Selection
    2. 3.2 Model Training and Compilation
  7. 4Application Development
    1. 4.1 System Flow
    2. 4.2 Object Tracker
    3. 4.3 Dashboard and Bounding Boxes Drawing
    4. 4.4 Physical Demo Setup
  8. 5Performance Analysis
    1. 5.1 System Accuracy
    2. 5.2 Frame Rate
    3. 5.3 Cores Utilization
    4. 5.4 Power Consumption
  9. 6Summary
  10. 7References

2.4 Data Augmentation

Data augmentation in machine learning includes generating altered copies of the pictures in a dataset. Data augmentation can be as simple as adding noise to the dataset in order to generate enough variation to prevent model overfitting. It is also used to expand the dataset by adding those altered copies. Two geometrical augmentation methods are applied in this demo: flip right-left and rotation. First flipped copies are created for each picture, which brings the total number of pictures to 400x2=800. Then five rotated copies of each picture are created, which brings the total number of pictures up to 800+800x5=4800 pictures. The rotation angle is randomly selected for each picture. This step substantially increases the total number of pictures in the dataset without the burden of data capturing and annotation.