TIME-AWARE AUTO WHITE BALANCE IN MOBILE PHOTOGRAPHY

Smart overview of the Invention

The patent application describes a method for enhancing auto white balance (AWB) in mobile photography by incorporating time-awareness. Traditionally, AWB adjusts the color balance of images to correct undesirable color casts due to lighting conditions. This method introduces a novel approach by integrating time-capture features derived from image metadata, alongside histogram features from the raw image, to improve illuminant estimation.

Technical Approach

The method begins with obtaining a raw image and its associated metadata. A time-capture feature is generated from this metadata, while a histogram feature is derived from the raw image data. These features are processed through a neural network to determine the illuminant chromaticity. This chromaticity is then converted into an RGB illuminant color, which is used to adjust the white balance parameters of an Image Signal Processor (ISP).

Device Implementation

An electronic device implementing this method includes memory to store instructions and at least one processor to execute them. The processor performs tasks such as obtaining the raw image and metadata, generating the necessary features, and using a neural network to determine the illuminant chromaticity. The device then converts this chromaticity into an RGB color to adjust the ISP's white balance settings.

System and Methodology

The system involves a combination of hardware and software components. The hardware includes specialized control units that execute the described methods, while the software involves algorithms that process image data and metadata. The methodology emphasizes flexibility, allowing various components and features to be combined or modified to suit different implementations and use cases.

Additional Information

The application also includes diagrams and flowcharts illustrating the environment and processes used in this method. These visual aids demonstrate the system's architecture and the step-by-step operations involved in time-aware auto white balancing. Furthermore, the application highlights the potential for integrating existing illuminant estimation modules, enhancing the system's adaptability and efficiency.