3D Halftoning

3D printing is becoming increasingly popular around the world today. By utilizing 3D printing technology, customized products can be manufactured much more quickly and efficiently with much less cost. However, 3D printing still suffers from low quality surface reproduction compared with 2D printing. One effective approach to improve it is to develop an advanced halftoning algorithm for 3D printing. In this paper, a novel 3D DBS (Direct Binary Search) halftoning algorithm that can cooperate with current 3D printing technology is proposed which can generate high quality surface reproduction. Introduction The emerging 3D printing technology enables human to create complicated 3D objects and structures with different kinds of materials more economically and efficiently than the traditional manufacturing process. However, currently more attention has been paid on the development of 3D printing process, material and structure rather than the 3D object surface image reproduction. With the growing need for high surface printing quality, improving the image reproduction quality on 3D surface has become more and more important in 3D printing area. A number of 3D printing techniques that are able to do 3D color printing have been developed for some time. One technique uses powder and binder to generate 3D objects. It spreads thin layers of powders across a platform and the print head selectively deposits color binders onto the powder. And finally the redundant powders will be removed. Another technology utilizes layerlaminating system to do 3D color printing. It uses normal ink jet printing technology to print a slice of the 3D model on a piece of standard copy paper and then stack them up. A cutting blade traces the outline of the model to remove the excess paper. However, most of them suffer from low resolution and rough surface. Recently the *Research supported by HP Inc., Palo Alto, CA. Multi-Jet technology from companies like HP Inc. has been revealed to public. This technology enables micro manipulation of a single voxel, which can produce much smoother and higher resolution surface with high quality image reproduction. 3D printers equipped with this technology will have more potential in generating high quality and complex appearance 3D objects. By utilizing Multi-Jet technology which enables voxel-wise printing, digital halftoning will be able to play a core role in improving 3D surface reproduction quality. In general, digital halftoning algorithms can be classified into three categories: screening, error diffusion and search-based methods. Screening only requires point-topoint memoryless comparison with a threshold to generate a halftone image with binary values. Error diffusion method includes point-to-point comparison also and requires a neighborhood processing with a little bit memory. Search-based method directly searches for the best arrangement of binary pattern for the halftone image by minimizing the error metric of perceived error. Search based methods are usually iterative which requires more computation than screening and error diffusion methods which require only one pass of the entire image and no or very small memory. However, search-based methods yield better halftone quality than screening and error diffusion at the cost of higher computation. Among several reported search-based methods, DBS which was first reported in [1] yields very good halftone quality and is currently one of the most widely used search-based methods. In fact, several 3D digital halftoning algorithms have been reported recently based on the 2D halftoning algorithms of screening and error diffusion. For example, the monochrome 3D dispersed-dot screening digital halftoning algorithm is reported in [2]. It generated a 3D volume threshold matrix to halftone a 3D object. Some other works like Brunton et al.[3] and Zhou et al. [4] applied error diffusion to layer based 3D object halftoning for color and monochrome 3D halftoning respectively. IS&T International Symposium on Electronic Imaging 2017 Color Imaging XXII: Displaying, Processing, Hardcopy, and Applications 147 https://doi.org/10.2352/ISSN.2470-1173.2017.18.COLOR-048 © 2017, Society for Imaging Science and Technology