Research on Visual Evaluation of Video Images
We are carrying our projects to create a framework under which digital video content can be shown with the color characteristics intended by the creator regardless of where and by whom it is shown. Through our studies, we have gained valuable knowledge and insight into various methods and standards of evaluating content quality.
We have also conducted experiments toward the future formulation of a color reproduction standard for video images.
By displaying projected video content and evaluating the impressions of viewers, we have come to recognize the possibility of reproducing colors that give similar impressions even when displayed by different methods.
Furthermore, we have closely examined the current state of image quality evaluation by analyzing content elements and methods of image quality evaluation. Based on viewers’ impressions of content, we also examined possible approaches, actual examples, and future directions regarding ways of evaluating content. We studied words used to evaluate content in a wide range of genres, the relationship between psychological and emotional factors in evaluating images, and the connection between visual stimulation and the brain. Through these efforts, we found possible methods of evaluating video images based on viewers’ impressions.
As for the future, we intend to examine more accurate evaluation indicators and quantitative evaluation methods through experiments. We also plan to create high-quality content for use in evaluation and to study quantitative color adjustment methods.
3D Content Research
In pace with the increasing sophistication and diversification of digital content made possible by dramatic advances in information technology and digital devices, more efforts are being made to create 3D content with highly realistic images.
Although expectations for the use of 3D content are growing in many fields, the reality is that 3D has not yet become an effective means of the digital content industry promotion.
In our 3D content research, we extracted vital information and closely looked into challenges in production and utilization of 3D content. Through this, we were able to clarify effective methods and crucial elements to be examined toward future industry promotions.
Project to Promote Supersensitive Video Shooting Systems
In academic and other circles, there are high expectations toward supersensitive video shooting systems that clearly capture images of natural phenomena and other events even in low-light conditions. One of the few supersensitive video technologies developed in Japan is the HARP (High-gain Avalanche Rushing amorphous Photoconductor) imaging device, which will likely have various applications in a range of academic and industrial fields.
In our Promotion of Supersensitive Video Shooting Systems, we sought areas in which such systems could be effectively used such as broadcasdting, security/monitoring, medicine, academics, environmental assessment and entertainment.
During development of the system, we improved its sensitivity to red and explored the possibility of enhancing its sensitivity to green. We also added a signal processing function and redesigned the operation panel to reduce color afterimage and raise camera functionality while also improving the camera's picture quality and operability.
Images obtained during test (illuminance: 5lx, F2.0)
HARP camera with boosted red sensitivity
CCD camera (+6dB)
Red in the dolls clothing is captured more vividly thanks to higher sensitivity to red. HARP’s ability to reproduce gradations even in dark areas resulted in a clearly reproduced doll shadow on the black wall.