Stereoscopic research on videos and images has been a new trend over an extended duration. For various applications of stereoscopic videos and images, visual quality has been fundamental. Quality assessment of images and video processing is essential. It is useful for companies that produce quality images or videos.

Unfortunately, humans lack the knowledge needed to create the best stereoscopic images. In this paper, you will discover the results of the research done to measure human vision judgement. The database obtained from the analysis of quality stereoscopic images can help to improve the perception of stereoscopic images.

Image Quality

In recent years, autostereoscopic displays have developed tremendously. Research on video processing and stereoscopic images has become a new trend. As a result of this research, people have developed an interest in standardizing and designing technologies. You can use this imaging technology for viewing 3DTV content, storage and production of quality images.

In your research, you’ll understand specific content about a technique called depth-image-based rendering (DIBR). For standard displays, maps can be used to produce stereoscopic images. When conducting research programs, you must focus on specific issues such as:

• How image quality relates to distorted parameters

• The relationship between distorted images perception and index of objective measures

Experimental Method

For research-based on human trial, you can assess the quality stereoscopic images using the recommendations for ITU-R. In this method, a researcher views a pair of photographs of a similar model. They comprise of uncompressed and compressed original photos. You calculate the mean score for every test on representation and condition following all combinations required.

  • Viewing Conditions and Display: in the research, you put on polarized glasses to separate left and right images. This research should be conducted in a dark room with minimal light.
  • Database for Stereoscopic Images: This database is derived from images that reflects the diversity of all image contents.