Image signal processors (ISP) plays a significant role in camera systems by converting the RAW image from image sensor to a processed image. In order to achieve best image quality, the ISP parameters have to be configured in an iterative manner for various lighting conditions and scenarios, which is carried out by a camera tuning engineer. Usually, the manual tuning process takes up to several weeks to months due to huge number of ISP parameters to be optimized and the iterations involved to achieve good image quality. In this paper, we present a novel approach to automatically tune ISP parameters based on a multi-stage multi-criteria optimization approach using Non sorted Genetic algorithm (NSGA-II) for achieving objective and subjective image quality. In this approach, we focus on important blocks in ISP such as noise reduction, sharpness and tone mapping for human vision use-cases for camera systems widely used for smart phones or smart home IoT devices. The experiments for validating our approach are carried out under different scenarios using Qualcomm’s Spectra 380 ISP simulator and OV13880 sensor and the performance of automatic tuned IQ is compared with manual tuned IQ and some of the previous works done for automatically tuning ISP parameters. With the automatic ISP tuning approach, we verify the significant performance improvement in terms of IQ metrics and time consumed for the tuning process when compared to manual tuning approach.
In this work, we present a psychophysical study, in which, we analyzed the perceptual quality of images enhanced with several types of enhancement algorithms, including color, sharpness, histogram, and contrast enhancements. To estimate and compare the qualities of enhanced images, we performed a psychophysical experiment with 35 source images, obtained from publicly available databases. More specifically, we used images from the Challenge Database, the CSIQ database, and the TID2013 database. To generate the test sequences, we used 12 different image enhancement algorithms, generating a dataset with a total of 455 images. We used a Double Stimulus Continuous Quality Scale (DSCQS) experimental methodology, with a between-subjects approach where each subject scored a subset of the total database to avoid fatigue. Given the high number of test images, we designed a crowd-sourcing interface to perform an online psychophysical experiment. This type of interface has the advantage of making it possible to collect data from many participants. We also performed an experiment in a controlled laboratory environment and compared its results with the crowd-sourcing results. Since there are very few quality enhancement databases available in the literature, this works represents a contribution to the area of image quality.