JeVois  1.4
JeVois Smart Embedded Machine Vision Toolkit
Share this page:
Changes and new features in JeVois 1.4

JeVois 1.4 brings the following new features:

  • New /jevois/share/ directory to store data that may be shared across several machine vision modules, such as neural network config files and weights, pre-trained classifiers, etc
  • Object recognition on ImageNet (1000 object categories) and Pascal VOC using Darknet and Darknet YOLO. Not fully optimized for speed yet but already looking very promising.
    darknet-demo.png
    See two new modules in jevoisbase:
    • DarknetSingle passes a central 224x224 crop to a Darknet neural network. The neural network precicts the top N object category candidates (among 1000 possible from imagenet), and those above a minimum confidence threshold are shown and sent to serial.
    • DarknetYOLO passes a whole 640x480 frame to a Darknet YOLO (you only look once) network that combines detection (finding where objects might be) and identification (what objects are) for all objects in the scene in one pass.

  • A few more image conversion functions in RawImageOps.H
  • New machine vision modules for ARtoolkit (ultra fast detection of augmented reality markers), Darknet, Darknet YOLO.
  • Updated RoadNavigation module to output correct VP information.
  • New tutorial on how to build a simple vision-guided mobile robot