Flexible Storage of Images for Digital Cameras

Currently, digital still image cameras offer the user little flexibility. After initially selecting the number of images to be stored (from a limited number of possible settings), this selection cannot easily be changed at a later time. Furthermore, when only a few pictures have been taken, a large part of the storage capacity may remain unused (whereas it could have been used to improve image quality). Finally, allocating a fixed amount of storage to each image causes the image quality to vary: more complex images will be stored at a lower than average quality. To overcome these three limitations, we propose a new storage method that efficiently uses all of the available storage capacity and offers the user maximal flexibility. It is based on the application of bit-rate scalable image compression methods. These scalable compression methods have the feature that the bit string that describes the image can be cut off at any point, while the bits that are not cut off always provide the best possible reconstruction quality for the image, given the number of bits used to represent it. This feature allows us to cut the encoded bit string into a number of equal-sized blocks and drop blocks from the end of the string. Scalable image compression further has the property that the bits that occur first in the compressed image string provide a larger reduction of the distortion in the reproduction made by the decoder than the bits that occur later in the bit string. Therefore, a block of compressed data from the beginning of the bit string is more important for the image quality than a block of data from the end of the string. By computing the quality contribution for each block, we can compare the blocks of each new image to be stored with the already stored blocks of previous images. The previously stored blocks can be overwritten by new blocks that have a higher quality contribution. Our storage method fully uses the available storage capacity to store the images at the highest overall quality. Initially, when only a few pictures have been taken, they are stored losslessly (or near-losslessly). When more images have to be added, the storage space occupied by the previously stored images is adjusted optimally, in the rate-distortion sense, to fit in each new image. Thus, a variable amount of space is allocated to each image, depending on its complexity. The method allows the selected data of each new image to simply overwrite part of the data of one or more previous images, so it operates without any decoding and re-encoding of image data on the storage medium.