3D shape reconstruction from images is an active topic in computer vision. Shape-from-Focus (SFF) is an important approach which requires image stack in a focus controlled manner to infer the 3D shape. In this article, 3D reconstruction of synthetic gastrointestinal regions is done using SFF. Image stack is generated in Blender software with focus controlled camera. A color focus measure is applied for shape recovery followed by a weighted L2 regularizer to estimate for inaccurate depth values. A precise comparison is done between recovered shape and ground truth data by measuring the depth error and correlation between them. Results shows that SFF technique will be practical for 3D reconstruction of GI regions with focus and motion controlled pillcams which is technologically feasible to implement.
Publisher’s Note: This paper, originally published on 13 March 2015, was replaced with a corrected/revised version on 6 August 2019. If you downloaded the original PDF but are unable to access the revision, please contact SPIE Digital Library Customer Service for assistance.
Surgical robot technology has revolutionized surgery toward a safer laparoscopic surgery and ideally been suited for surgeries requiring minimal invasiveness. Sematic segmentation from robot-assisted surgery videos is an essential task in many computer-assisted robotic surgical systems. Some of the applications include instrument detection, tracking and pose estimation. Usually, the left and right frames from the stereoscopic surgical instrument are used for semantic segmentation independently from each other. However, this approach is prone to poor segmentation since the stereo frames are not integrated for accurate estimation of the surgical scene. To cope with this problem, we proposed a multi encoder and single decoder convolutional neural network named StreoScenNet which exploits the left and right frames of the stereoscopic surgical system. The proposed architecture consists of multiple ResNet encoder blocks and a stacked convolutional decoder network connected with a novel sum-skip connection. The input to the network is a set of left and right frames and the output is a mask of the segmented regions for the left frame. It is trained end-to-end and the segmentation is achieved without the need of any pre- or post-processing. We compare the proposed architectures against state-of-the-art fully convolutional networks. We validate our methods using existing benchmark datasets that includes robotic instruments as well as anatomical objects and non-robotic surgical instruments. Compared with the previous instrument segmentation methods, our approach achieves a significant improved Dice similarity coefficient.
This paper presents an image-based method to measure reflectance of a homogeneous flexible object material (usually used in packaging). A point light source and a commercially available RGB camera is used to illuminate and measure the radiance reflected from the object surface in multiple reflection directions. By curving the flexible object onto a cylinder of known radius we are able to record radiance at multiple reflection angles in a faster way. In order to estimate the reflectance and to characterise the material, a spectralon reference tile is used. The spectralon tile is assumed to be homogenous and has near lambertain surface properties. Using Lambert’s cosine law, irradiance at a given point on the object surface is calculated. This information is then used to calculate a BRDF using Phong reflection model to describe the sample surface reflection properties. The measurement setup is described and discussed in this paper along with its use to estimate a BRDF for a given material/substrate. Results obtained indicate that the proposed image-based technique works well to measure light reflected at different planar angles and record information to estimate the BRDF of the sample materials that can be modelled using Phong reflection model. The object material properties, sample curvature and camera resolution decides the number of incident and reflection angles at which the bi-directional reflectance, or the material BRDF, can be estimated using this method.
Color deficient people might be confronted with minor difficulties when navigating through daily life, for example when reading websites or media, navigating with maps, retrieving information from public transport schedules and others. Color deficiency simulation and daltonization methods have been proposed to better understand problems of color deficient individuals and to improve color displays for their use. However, it remains unclear whether these color prosthetic" methods really work and how well they improve the performance of color deficient individuals. We introduce here two methods to evaluate color deficiency simulation and daltonization methods based on behavioral experiments that are widely used in the field of psychology. Firstly, we propose a Sample-to-Match Simulation Evaluation Method (SaMSEM); secondly, we propose a Visual Search Daltonization Evaluation Method (ViSDEM). Both methods can be used to validate and allow the generalization of the simulation and daltonization methods related to color deficiency. We showed that both the response times (RT) and the accuracy of SaMSEM can be used as an indicator of the success of color deficiency simulation methods and that performance in the ViSDEM can be used as an indicator for the efficacy of color deficiency daltonization methods. In future work, we will include comparison and analysis of different color deficiency simulation and daltonization methods with the help of SaMSEM and ViSDEM.
KEYWORDS: RGB color model, Image quality, Visualization, Image enhancement, Visual system, Lead, Image processing, MATLAB, Information visualization, Human vision and color perception
Color-deficient observers are often confronted with problems in daily life due to the fact that some colors appear
less differentiable than for normal sighted people. So-called daltonization methods have been proposed to increase
color contrast for color-deficient people. We propose two methods for better daltonization solutions by Spatial
Intensity Channel Replacement Daltonization (SIChaRDa). We propose replacing the intensity channel with
a grayscale version of the image computed by using spatial color-to-gray methods that are either capable of
translating color contrasts into lightness contrasts or that are capable of translating color edges into lightness
edges, and/or integrating information from the red–green channel into the intensity channel. We tested two
implementations on different types of images, and we could show that results depend on the one hand on the
algorithm used for computing the grayscale image, and on the other hand on the content of the image. We show
that the spatial methods work best on real-life images were confusing colors are directly adjacent to each other,
respectively where they are in close proximity. On the contrary, using composed artificial images with borders
of white space between colors – like for example in the Ishihara plates – leads only to unsatisfactory results.
QuickEval is a web application for carrying out psychometric scaling experiments. It offers the possibility of
running controlled experiments in a laboratory, or large scale experiment over the web for people all over the
world. It is a unique one of a kind web application, and it is a software needed in the image quality field.
It is also, to the best of knowledge, the first software that supports the three most common scaling methods;
paired comparison, rank order, and category judgement. It is also the first software to support rank order.
Hopefully, a side effect of this newly created software is that it will lower the threshold to perform psychometric
experiments, improve the quality of the experiments being carried out, make it easier to reproduce experiments,
and increase research on image quality both in academia and industry. The web application is available at
www.colourlab.no/quickeval.
The original presentation of Retinex, a spatial color correction and image enhancement algorithm modeling the human vision system, as proposed by Land and McCann in 1964, uses paths to explore the image in search of a local reference white point. The interesting results of this algorithm have led to the development of many versions of Retinex. They follow the same principle but differ in the way they explore the image, with, for example, random paths, random samples, convolution masks, and variational formulations. We propose an alternative way to explore local properties of Retinex, replacing random paths by traces of a specialized swarm of termites. In presenting the spatial characteristics of the proposed method, we discuss differences in path exploration with other Retinex implementations. Experiments, results, and comparisons are presented to test the efficacy of the proposed Retinex implementation.
This paper describes a novel implementation of the Retinex algorithm with the exploration of the image done
by an ant swarm. In this case the purpose of the ant colony is not the optimization of some constraints but
is an alternative way to explore the image content as diffused as possible, with the possibility of tuning the
exploration parameters to the image content trying to better approach the Human Visual System behavior. For
this reason, we used "termites", instead of ants, to underline the idea of the eager exploration of the image.
The paper presents the spatial characteristics of locality and discusses differences in path exploration with other
Retinex implementations. Furthermore a psychophysical experiment has been carried out on eight images with
20 observers and results indicate that a termite swarm should investigate a particular region of an image to find
the local reference white.
The archives of motion pictures represent an important part of precious cultural heritage. Unfortunately, these
cinematography collections are vulnerable to different distortions such as colour fading which is beyond the
capability of photochemical restoration process. Spatial colour algorithms-Retinex and ACE provide helpful
tool in restoring strongly degraded colour films but, there are some challenges associated with these algorithms.
We present an automatic colour correction technique for digital colour restoration of strongly degraded movie
material. The method is based upon the existing STRESS algorithm. In order to cope with the problem of highly
correlated colour channels, we implemented a preprocessing step in which saturation enhancement is performed
in a PCA space. Spatial colour algorithms tend to emphasize all details in the images, including dust and
scratches. Surprisingly, we found that the presence of these defects does not affect the behaviour of the colour
correction algorithm. Although the STRESS algorithm is already in itself more efficient than traditional spatial
colour algorithms, it is still computationally expensive. To speed it up further, we went beyond the spatial
domain of the frames and extended the algorithm to the temporal domain. This way, we were able to achieve
an 80 percent reduction of the computational time compared to processing every single frame individually. We
performed two user experiments and found that the visual quality of the resulting frames was significantly better
than with existing methods. Thus, our method outperforms the existing ones in terms of both visual quality
and computational efficiency.
Gamut mapping algorithms are currently being developed to take advantage of the spatial information in an
image to improve the utilization of the destination gamut. These algorithms try to preserve the spatial information
between neighboring pixels in the image, such as edges and gradients, without sacrificing global contrast.
Experiments have shown that such algorithms can result in significantly improved reproduction of some images
compared with non-spatial methods. However, due to the spatial processing of images, they introduce unwanted
artifacts when used on certain types of images. In this paper we perform basic image analysis to predict whether
a spatial algorithm is likely to perform better or worse than a good, non-spatial algorithm. Our approach starts
by detecting the relative amount of areas in the image that are made up of uniformly colored pixels, as well
as the amount of areas that contain details in out-of-gamut areas. A weighted difference is computed from
these numbers, and we show that the result has a high correlation with the observed performance of the spatial
algorithm in a previously conducted psychophysical experiment.
A method is proposed for performing spectral gamut mapping, whereby spectral images can be altered to fit within an approximation of the spectral gamut of an output device. Principal component analysis (PCA) is performed on the spectral data, in order to reduce the dimensionality of the space in which the method is applied. The convex hull of the spectral device measurements in this space is computed, and the intersection between the gamut surface and a line from the center of the gamut towards the position of a given spectral
reflectance curve is found. By moving the spectra that are outside the spectral gamut towards the center until the gamut is encountered, a spectral gamut mapping algorithm is defined. The spectral gamut is visualized by approximating the intersection of the gamut and a 2-dimensional plane. The resulting outline is shown along with the center of the gamut and the position of a spectral reflectance curve. The spectral gamut mapping algorithm is applied to spectral data from the Macbeth Color Checker and test images, and initial results show that the amount of clipping increases with the number of dimensions used.
The spectral integrator at the University of Oslo consists of a lamp whose light is dispersed into a spectrum by means of a prism. Using a transmissive LCD panel controlled by a computer, certain fractions of the light in different parts of the spectrum is masked out. The remaining spectrum is integrated and the resulting colored light projected onto a dispersing plate. Attached to the computer is also a spectroradiometer measuring the projected light, thus making the spectral integrator a closed-loop system. One main challenge is the generation of stimuli of arbitrary spectral power distributions. We have solved this by means of a computational calibration routine: Vertical lines of pixels within the spectral window of the LCD panel are opened successively and the resulting spectral power distribution on the dispersing plate is measured. A similar procedure for the horizontal lines gives, under certain assumptions, the contribution from each opened pixel. Hereby, light of any spectral power distribution can be generated by means of a fast iterative heuristic search algorithm. The apparatus is convenient for research within the fields of color vision, color appearance modelling, multispectral color imaging, and spectral characterization of devices ranging from digital cameras to solar cell panels.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.