Blog_banner_beyond_the_datasheet

OCR algorithms work better with high quality images for accurate Automated License Plate Recognition ALPR

Posted by Gretchen Alper on Wed, Feb 15, 2012

Unlike what is shown on TV, you cannot zoom into a blurry image and expect to get more details.  An image with acceptable sharpness and contrast must be acquired with the appropriate system from the start.  This means the right image sensor, camera, optics, and lighting all combined in a reliable way.

So what defines good image quality for ALPR?

The first step is to have reliable triggering in order to have the license plate in the proper location in the image, which can be especially difficult in multi-lane systems.  After that, a good/accurate image can be described by:

  • Good Sharpness
  • Sufficient Contrast
  • Free of artifacts
  • And sometimes with accurate color

These are qualitative explanations so here some images to demonstrate the point.

OCR image for ALPR

Figure 1.  Artifacts from insufficient lighting control

 

ITS ALPR Motion Blur

Figure 2.  Insufficient Sharpness due to Motion Blur

Low contrast ITS ALPR

Figure 3.  Insufficient Contrast from limited dynamic range

 

Some good background information on license plate acquisition algorithms and technology is provided on: http://www.platerecognition.info/1102.htm

The sources of these image quality issues can vary.  Some possible reasons are shown in the table below and are further detailed in our next series of blogs.

Image Quality Parameter

Corresponding Source of Limitations

Image System Parameters to Control

Sharpness

Limited Depth of Field

Motion Blur

Variable Lighting

 

F Value of Lens

Sensitivity of Image Sensor

Iris Control

 

Contrast

Limited number of images

Reflections of the license plate

Reflections of snow, rain, flog

Frame rate of the image sensor/camera

Dynamic range of the image sensor/camera

Artifacts

Ghost images

Bright spots and streaks from sun exposure and reflections

Alignment of filter, lens, and lighting

Channel Matching in the camera

Blooming and smear control in the camera

Color

Inaccurate Color reproduction

Accurate color calculations and automatic white balance in the camera

Proper alignment of the entire optical path determines the quality of the image captured, which is especially vital in high-speed situations.  With a higher quality of the input image, the better starting point for the license plate recognition algorithm, and therefore the higher license plate recognition accuracy. 

For more details on optimizing sharpness and contrast, and minimizing artifacts, see our blogs in the coming weeks.

Can’t wait?

You can already read the following blogs about this:

 

 

Topics: Applications

Previous blog:

Next blog: