Method of modeling and interactively visualizing the appearance of a surface
Background of the Invention
Field of the Invention
The present invention relates to a method of modeling and interactively visualizing directed reflectance data, and generally directed radiance data. More precisely, the invention is concerned with processing and interactively displaying directed reflectance or directed radiance data at a high degree of fidelity in the context of visual perception. The directed reflectance data is not limited to color information. On the contrary, subjectively important directed reflectance characteristics like gloss behavior is within the scope of the invention; yielding a system that based on interactivity facilitates visualization of multiple characterizing appearance aspects of a surface. The directed reflectance data may by obtained by measuring directed reflec- tance from physical surfaces or be modeled or simulated.
Description of Related Art
The directed reflectance properties of surfaces such as printed-paper surfaces are regularly subject to evaluation in order to improve the subjective ap- pearance of the surfaces. A typical evaluation process is performed by a group of persons each visually inspecting physical sample surfaces, inter alia so called tone plates, and giving marks on a scale to the quality of the surfaces from certain aspects such as gloss.
The sample surfaces are very sensitive to inappropriate handling — a single finger print mark on the viewing surface will in the case of gloss evaluation completely destroy the sample making it impossible to further evaluate that sample. Also for other types of print quality evaluations the sample surfaces are sensitive to improper handling. As the samples are handled manually, it is further very difficult to have uniform viewing conditions, such as viewing angle and illumination, for all per- sons in the group. Another common practice when subjectively evaluating samples is to use the pair-wise comparison approach. The main idea is to view and judge all possible pairs of samples, given the whole set of samples evaluated. It is then crucial that the two samples are handled and viewed identically, if the judgment is to be correct.
Objects of the Invention
An object of the invention is to provide a method for computerized mod- eling and interactive visualization of the directed reflectance properties of a surface making it possible to visually and interactively present a measured, modeled or simulated surface via a graphic computer interface. By interactively visualizing the surface via an interface, the three mentioned problems are dealt with. First, the computer- based interactive visualization implies that the handling of the physical specimen is reduced to a single measurement session, which can further be performed by skilled and experienced personnel, minimizing the risk of destroying the physical specimen. Second, the computer-based interactive visualization facilitates an exact control of high resolution of the sample handling, and is thus a powerful tool as changes in the appearance of gloss often can be drastic even for limited changes in the illumination- sample-observer set-up. Third, the computer-based interactive visualization also guarantees a pair-wise comparison performed under identical inspection conditions, such as tilting or bending of the two surfaces. The total control of the inspection conditions is essential for a well-performed pair-wise evaluation, and it is crucial if the surfaces to be evaluated are very similar. Another object is to interactively manipulate the orientation and geometry of the sample surface by the computer interface via an also visualized virtual sample holder given a neutral gloss behavior. The visualized virtual sample holder is an aid for the observer to gain information about the present shape of the sample surface and thus an aid to perceive and evaluate multiple characterizing aspects of the sample surface visualized. By using an interactive visual interface, in combination with the visualized virtual sample holder, a high degree of the state of "Presence" is achieved for the evaluator. The concept of "Presence" is frequently used in the area of Virtual Reality as to describe the state of the subject being part of the simulated reality, rather than being an observer of a simulated scene. An important constituent in presence for the human eye is believed to be the sense of interaction that is experienced when manipulating a visual interface.
Summary of the invention
According to one aspect of the invention there is provided a method of modeling and visualizing the appearance of a surface, comprising: providing a directed radiance model of an irradiated surface, compris- ing:
providing a spectrally resolved irradiance model comprising energy distribution as a function of spectral wavelength;
dividing the surface in a plurality of part-surfaces, providing spatial resolution; and
determining spectrally resolved directed radiance values of each part-surface in an interval of viewing angles, providing spectral and angular resolution;
selecting sets of directed radiance values among the determined spectrally resolved directed radiance values, that simulates certain viewing angles of the irradiated surface;
presenting the sets of directed radiance values visually as images of the irradiated surface by a graphic computer interface; and
manipulation, dynamically and interactively, of the images in the graphic computer interface to thereby simulate manipulation of the irradiated physical surface.
The dynamic and interactive manipulation of the images in the graphic computer interface is preferably realized via manipulation, dynamically and interactively of a simulated sample holder of the irradiated surface.
The manipulation is also preferably selected from a group of actions of a virtual sample holder of the irradiated surface comprising:
inclination of the sample holder, translation of the sample surface over the sample holder, flexing of the sample holder, and topographic reshaping of the sample holder.
The model of the sample surface may optionally be obtained by registering directed reflectance data from a real sample surface or obtained by simulating directed reflectance data from a virtual sample surface.
Brief Description of the Drawings
FIG. 1 is a diagrammatic view of a typical viewing situation to be visualized and modeled by the invention;
FIG. 2 shows a setup for obtaining directed reflectance data as to further modeling a sample surface from an actual sample surface; FIG. 3 is a diagrammatic representation of an information volume;
FIG. 4A shows a "screen shot" of a computer graphic interactive interface for dynamic visualization and manipulation of a sample surface model, here exemplified by three different sample surfaces, all printed in black full tone;
FIG. 4B shows a "screen shot" of the same computer graphic interactive inter- face, and exemplified by the same three different sample surfaces, as in FIG. 4A, but here also a digital image information is visualized together with the local angle and reflectance information; and
FIG. 5 is a diagram showing a comparison between samples first measured and then visualized and judged on a computer screen vs. the results of the evalua- tion of the same real (physical) samples.
Detailed Description of the Preferred Embodiment
In the virtual viewing situation illustrated in FIG. 1 to be modeled and simulated by the invention, a surface 12 of a sample 10 is illuminated by a light source 16 at an angle A to the surface normal N. The illumination may be realized by a so- called Standardized llluminant, i.e. an illuminant with a well-defined energy distribution as a function of spectral wavelength and that meets a certain given standard specification. The frequent use of fluorescent agents in e.g. printing paper, might proof it necessary to generalize this description to include other incidence radiation
also than the visible light only. The characterizing feature of fluorescent agents is the ability to transform the electromagnetic radiation from shorter to longer wavelengths. The most common use of fluorescent agents for printing papers is to transform ultraviolet (invisible) irradiation into (visible) light radiation and hence achieving an in- crease of the perceived whiteness of the product.
The characterization necessary when using fluorescent agents, is to take also the directed emitted radiation from each part-surface into consideration. The characterization can be done either directly as radiation values or relative to a reference surface (a perfect diffusor) as a radiation factor. The generalization from the directed reflectance domain into the more general radiant domain is well within the capabilities of the invention. For many applications the directed reflectance characterization is however sufficient and as this term is more familiar, the term "directed reflectance" is frequently used in the describing text, bearing in mind the possibility of the generalization into directed radiant and directed radiance characterizations. Reflected light from light source 16 is viewed from a variable angle Zl to the surface normal N and sensed and registered by a viewer 18.
As indicated in the circular enlarged area of FIG. 1, in the preferred embodiment, the surface 12, for modeling purposes, is considered as divided in a large number of part-surfaces 14, where each part surface is considered to have uniform directed reflectance properties over its area. The location of each part-surface on the sample surface 12 is determined by its respective coordinates XI, Yl.
In the preferred embodiment, the virtual surface 12 is modeled from a real surface 12' of a sample by a registering apparatus 20 shown in FIG. 2. For a closer description of a similar apparatus delimited to the use of registering gloss only, as a special instance of the concept of directed reflectance, it is referred to Applicant's U.S. Patent 6,147,750 which is hereby incorporated as reference.
In the registering apparatus 20 the surface 12' is also radiated by the electromagnetic source 16, while the viewer 18' is a sensor such as a spatially and spectrally resolving receptor. A driving mechanism 22 rotates the surface 12' in a viewing area of the viewer 18'. A computer 24 is arranged to register, organize and process the directed reflectance value of each part-surface which is sensed and determined by the viewer 18'. More precisely, all directed reflectance values from the surface for all enforced inclinations ZN of the sample holder, and for all locations XN, YN of all part-surfaces 14 are registered in a memory of the computer and compiled as a di-
rected reflectance information-volume R = f (X, Y, Z) in a database, where R represents the measured spectrally resolved electromagnetic reflection in the coordinate (X, Y, Z). The X-Y-values represent the position within the sample surface and the Z- values represent the enforced sample holder inclination. The reflection may possibly be registered only by a limited set of characterizing spectrally selective basis functions, using e.g. the RGB-basis. In other words, a wide range of the whole range of possible viewing situations sensed by viewer 18' is stored in the information volume.
An information volume, which is well-known in the art of image and information processing, is diagrammatϊcally shown in FIG. 3 where the indicated plane 30 repre- sents the reflection from all part-surfaces at a zero enforced inclination.
In the computer graphic interactive interface 40 of FIG. 4A, there is shown an image area 42 for displaying the simulated appearance of one or more simulated or measured and manipulated surfaces contained in a respective image volume. More precisely, in the image area are visually presented sets of spectrally resolved di- rected reflectance values as images of the irradiated surface. The sets of spectrally resolved directed reflectance values are selected from the information volume to simulate certain viewing angles of the surface 12'. In the example of FIG. 4A, three surfaces 44, 46 and 48 as follows are shown:
Measured Paper Surface 44
LWC, Heat-Set-Offset paper (Not a typical LWC paper). An illustrative example of bad gloss quality.
Measured Paper Surface 46 Wood free, clay coated (~20g/m2) sheet fed offset.
Simulated Paper Surface 48 Directed reflectance: Mean gloss [0...255]: 60 Directed reflectance variation, band limited (large scale) [0...]: 3 Directed reflectance variation: Gauss(0, 0.05)
Surface topography:
Topological variation, band limited (large scale) [0...1]: 0.4
Topological variation: Gauss(0, 0.01)
The interface area 50 is a virtual sample holder given a neutral gloss behavior, which gives the user a visual indication of the actual orientation and shape of the manipulated surface. Generally referenced by 52 is an input area having slide bars 54 to control the orientation and shape of the virtual sample holder, i.e. to control and manipulate the selection of sets of spectrally resolved directed reflectance values. An obvious modification of the invention is to change the interface by omitting the slide bars 54 and thus to control the sample holder in a more direct form. This could be implemented via a computer input device, e.g. a mouse, or a virtual reality device like a "VR-glove" which is able to track and feed the computer with positions and movements.
In FIG. 4B, the same structural information of the sample surfaces used in FIG 4A, is visualized, but here also a digital image is super-positioned on the struc- tural information. The digital image information is a sub area of an image referenced by "N1.tif on "ISO 12640 Graphic technology - Prepress digital data exchange - Standard colour image data (SCID) - Annex A, Standard colour image digital data"
As is apparent from the above, in an alternative of the preferred embodiment, a simulated surface such as the simulated surface 48 may also be mod- eled and organized in an information volume. The simulated surfaces may, for example, be given subjectively varying good and bad directed reflectance qualities, and be presented as examples together with measured real surfaces in a print quality evaluation performed by a group of persons using the invention. A simulated surface may be modeled to more or less resemble a measured surface by systematic modifi- cation of the directed reflectance values in the information volume of the measured surface.
Example
One strong indication of the potential of the invention in terms of achiev- ing Presence during interactive visualization is based on results from two subjective evaluation studies. The first study was performed in a Master's Thesis work: Mikael Lindstrand, "A conceptual approach to describe gloss variation in printing paper" LiTH-ISY-EX-1624, University of Linkoping 1996, which is hereby incorporated as reference.
One conclusion based on results of a subjective evaluation of a set of physical sample surfaces, was that the performance of the individual panel members was very much dependent on previous experience in the field. All experienced members of the evaluation panel gave highly correlated results while the un-experienced members showed in general a very low correlation to other panel members. In order to test the possibility to use the invention as a tool and aid for subjective evaluation another test was recently performed. Based on the result from the previous subjective evaluation, only the experienced evaluators was chosen to participate in following subjective evaluation, where the samples were evaluated by using an implemen- tation of the invention. To summarize the results and experience from the two subjective evaluations: not only were the results of the evaluation based on the invention highly correlated to results of the evaluation for the physical samples, very interestingly the inter-evaluator variation was also even lower for the evaluation based on the invention. See FIG. 5 for a comparison between the ranking results of the evaluation of the samples first measured and then visualized and judged on a computer screen vs. the results of the evaluation of the same real (physical) samples.
Fig 5 illustrates both the similarities in the ranking result and relative lower degree of variation for the evaluation performed on the screen. The error bars illustrate a 95-percentic confidence interval. An important fact to stress is that the sample series was considered to be a challenge to evaluate, as the differences between samples were so small. This might be the reason why the inexperienced panel members failed to show results in consensus with the experienced evaluators.
To have evaluation tools that help reduce the variation in the results from the panel members is of high industrial value.