A Method for the Global Photographic Clarity based on Multifocal Pre-treatment of Partitioned View

As significant as shooting speed, clarity has long been TOP target in the realm of photography. To gain better clarity, the innovative automatic focusing (AF) technology has developed a complete new time for camera. Meanwhile, along with the boom of modern camera on mobile phone, better clarity within limited space for lens throws out higher requirement to shooting devices and algorithms. This paper introduces a new method of creating clearer photos based on creative mechanism of using AF and storage images. Inspired by the regional proposal of Region with Convolutional Neural Network (RCNN), the algorithm proposed in this paper uses partition to divide the whole target view (WTV) into small blocks with AF frame window so as to decrease the difficulty of getting a globally clear image. Then each of block would be detected with AF within the whole image when taking its central point as the focusing central point. When the focused data of the WTV was captured and temporarily stored, this solution will only save the part of the data where the clarity is the highest and label it with the block index i as “Best Field (BF)” i. After the similar processing of searching and saving BF(i) of all the blocks is finished, stitching all the BFs will produce the best image of the WTV. In sum, the solution advised in this paper innovates an ingenious and direct way to achieve the best clarity of the whole photo with AF and partition methods during taking picture, which not only produces much more clearer pixels and larger proportion of clear image for photographing the target scene with current sensor, but also avoid the involvement of manual work and eliminate the difficulty of professional operation required by image synthesis for creating globally clear photo.