- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2018 25th IEEE International Conference on Image Processing (ICIP) - Athens, Greece (2018.10.7-2018.10.10)] 2018 25th IEEE International Conference on Image Processing (ICIP) - Automatic ISP Image Quality Tuning Using Nonlinear Optimization
摘要： Image Signal Processor (ISP) comprises of various blocks to reconstruct image sensor raw data to final image consumed by human visual system or computer vision applications. Each block typically has many tuning parameters due to the complexity of the operation. These need to be hand tuned by Image Quality (IQ) experts, which takes considerable amount of time. In this paper, we present an automatic IQ tuning using nonlinear optimization and automatic reference generation algorithms. The proposed method can produce high quality IQ in minutes as compared with weeks of hand-tuned results by IQ experts. In addition, the proposed method can work with any algorithms without being aware of their specific implementation. It was found successful on multiple different processing blocks such as noise reduction, demosaic, and sharpening.
关键词： Image Quality Tuning,ISP,Nonlinear Optimization
更新于2025-09-10 09:29:36
-
Product-Specific Nonlinear Characterization and Instrument for Illumination Angle Measurement With Manufacturing Defects
摘要： Popular sensors that detect the quadrant of the illumination directions are often inadequate for advanced control systems, e.g., solar trackers and heliostats. We propose a theoretical and technical framework for some sensor-specific characterization tools together with an optical instrument, which makes possible the accurate measurements in terms of the illumination angle and its angular velocity at the presence of the manufacturing errors due to rough manufacturing processes (e.g., forging and casting). In particular, a parameter identification approach is employed to identify four uncertain incline angles of sensor products with the aid of an ingenious and controllable calibration device. The angular velocity of illumination is approximately derived through a one-step backward difference function together with finite impulse response filters to reject the disturbances. The experimental results show that the sensor-specific nonlinear characteristic functions improve the accuracy of the illumination angle measurement yielded by the general nonlinear characteristic functions for the ideal sensors up to 96%, where the measurement precision with 0.07° is achieved.
关键词： parameter estimation,optical sensor,characteristic function,nonlinear optimization,Calibration
更新于2025-09-04 15:30:14