If you are still wondering what the actual industrial applications for this object detection software could be, then take a look at the following:
Examples of industrial object detection
This is a quick overview of the applications that I used this software for to get accurate locations of detected objects:
- Dummy test cases containing rotation invariant detection of both cookies and candies on a set of different backgrounds.
- Automated detection and counting of microorganisms under a microscope instead of counting them yourself.
- Localization of strawberries for ripeness classification.
- Localization of road markings in an aerial imagery for automated creation of a GIS (Geographic Information System) based on the retrieved data.
- Rotation invariant detection of peppers (green, yellow, and red combined) on a conveyor belt combined with the detection of the stoke for effective robot gripping.
- Traffic sign detection for ADAS (Automated Driver Assist System) systems.
- Orchid detection for automated classification orchid species.
- Pedestrian detection and tracking in NIR images for security applications.
So, as you can see, the possibilities are endless! Now, try to come up with your own application and conquer the world with it.
Let's wrap up the chapter with a critical note on object detection using the Viola and Jones object categorization framework. As long as your application is focusing on detecting one or two object classes, then this approach works fairly well. However, once you want to tackle multiclass detection problems, it might be good to look for all the other object categorization techniques out there and find a more suitable one for your application, since running a ton of cascade classifiers on top of a single image will take forever.
Note
Some very promising object categorization frameworks that are in research focus at the moment, or that are a solid base for newer techniques, can be found below. They might be an interesting starting point for people wanting to go further than the OpenCV possibilities.
- Dollár P., Tu Z., Perona P., and Belongie S (2009, September), Integral Channel Features. In BMVC (Vol. 2, No. 3, p. 5)
- Dollár P., Appel R., Belongie S., and Perona P (2014), Fast feature pyramids for object detection. Pattern Analysis and Machine Intelligence, IEEE transactions on, 36(8), 1532-1545.
- Krizhevsky A., Sutskever I., and Hinton G. E (2012), Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems (pp. 1097-1105).
- Felzenszwalb P. F., Girshick R. B., McAllester D., and Ramanan D (2010), Object detection with discriminatively trained part-based models. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 32(9), 1627-1645.