Low-pass Image Research Articles

SU‐GG‐I‐93: A Fast and Simple Software‐Based Method for X‐Ray Scatter Reduction in Portable X‐Ray Systems

Purpose: To develop a processing method to reduce the effects of x‐ray scatter in images acquired with portable x‐ray systems without an anti‐scatter grid. Method and Materials: Due to the varying geometry inherent in portable x‐ray systems, the use of anti‐scatter grids with a high ratio is challenging. For clinical applicability, the algorithm must be fast and preserve the image texture. First an adequate understanding of the x‐ray scatter signal present in different radiography applications was obtained using Monte Carlo (MC) simulations of several clinical applications. From the simulations it was confirmed that the x‐ray scatter fields consisted of a low‐frequency offset, with very different shapes and magnitudes depending on the image being acquired. The developed algorithm, a modified version of unsharp masking, involves masking the open field area and the thin sections of the body, replacing these areas with the mean signal from the edges of the body, applying a low pass filter, and then subtracting a weighted version of the latter from the original image. The masking is performed to avoid the inclusion of the very high signal areas from the low pass image, which would result in an overestimation of the signal to be removed. Results: The algorithm reduced the x‐ray scatter signal in simulated MC images by 50&–80%, depending on the image and the location analyzed. When applied to images of a thorax phantom obtained with a portable x‐ray system without a grid, the quality of the images were improved enough to be comparable to those obtained with a clinical fixed system with a grid. Conclusion: The use of the developed scatter reduction algorithm seems to compensate for the inability to use a high ratio grid in portable systems. Further testing will be performed. Conflict of Interest: Research sponsored by GE Global Research.

Effect of Multiscale Processing in Digital Chest Radiography on Automated Detection of Lung Nodule with a Computer Assistance System

The aim of this study is to evaluate the effect of multiscale processing in digital chest radiography on automated detection of lung nodule with a computer-aided diagnosis (CAD) system. The study involved 58 small-nodule patient cases and 58 normal cases. The 58 patient cases included a total of 64 noncalcified lung nodules up to 15 mm in diameter. Each case underwent an examination with a digital radiography system (Digital Diagnost, Philips Medical Systems), and the acquired image was processed by the following three types of multiscale processing (Unique Image Processing Package, Philips Medical Systems) respectively: (1) standard image from the default processing parameter (structure preference, 0.0), (2) high-pass image with structure preference of 0.4, (3) low-pass image with structure preference of -0.4. The CAD output images were produced with a real-time computer assistance system (IQQAtrade mark-Chest, EDDA Technology). Two experienced chest radiologists established the nodule gold standard by consensus reading according to computed tomography results, and analyzed and recorded the detection of lung nodules and false-positive detections of these CAD output images. For the entire cases involved (each case with three types of different processing), a total of 348 observations were evaluated by the receiver operating characteristic (ROC) analysis. The mean area under the ROC curve (A ( z )) value was 0.700 for the standard images, 0.587 for the high-pass images, and 0.783 for the low-pass images. There were statistically significant A (z) values among these three types of processed images (p < 0.01). Multiscale processing in digital chest radiography can affect the automated detection of lung nodule by CAD, which is consistent with effects from visual inspection.

When structural noise is the signal: Speckle statistics in fluctuation electron microscopy

In a recent paper, Jungk et al. [Ultramicroscopy 104 (2005) 206–219] conclude that standard fluctuation electron microscopy speckle analysis, which examines the normalized variance of scattered intensity as a function of scattering angle, does not provide more information about disordered materials than does classical diffraction. In this letter, I point out flaws in both their experiments and their modeling that led them to this erroneous conclusion. In their experiments, they use resolutions that are excessively high relative to the length scale of the medium-range ordering in their samples, and consequently the diffraction fluctuations they seek are very weak. In their modeling, they attempt to correct this by applying low-pass image processing filters in order to broaden the effective point-spread function (i.e. to lower the image resolution after the experiment). Such filters, when applied to image intensities, do not emulate properly the evolution of coherent speckle as the point-spread function becomes wider. I affirm that fluctuation microscopy does indeed provide more information about disordered materials than classical diffraction when the experiments and modeling are conducted appropriately.

Transflective spatial filter based on azo-dye-doped cholesteric liquid crystal films

This work demonstrates the feasibility of exploiting the photoisomerization effect in azo-dye-doped cholesteric liquid crystal (DDCLC) films with a concomitant decline of the phase transition temperature from the cholesteric to an isotropic phase (TCh-I) as a spatial filter. The fabrication depends on the fact that the various intensities of the diffracted orders are responsible for the various degrees of transparency associated with the photoisomerized DDCLC film. High- and low-pass images in the Fourier optical signal process can be simultaneously observed via reflected and transmitted signals, respectively. A simulation is also performed, and the results are consistent closely with experimental data.

Individual differences in FFA activity suggest independent processing at different spatial scales

The brain processes images at different spatial scales, but it is unclear how far into the visual stream different scales remain segregated. Using functional magnetic resonance imaging, we found evidence that BOLD activity in the fusiform face area (FFA) reflects computations based on separate spatial frequency inputs. When subjects perform different tasks (attend location vs. identity; attend whole vs. parts) or the same task with different stimuli (upright or inverted) with high- and low-pass images of cars and faces, individual differences in the FFA in one condition are correlated with those in the other condition. However, FFA activity in response to low-pass stimuli is independent of its response to high-pass stimuli. These results suggest that spatial scales are not integrated before the FFA and that processing in this area could support the flexible use of different sources of information present in broad-pass images.

Automatic detection of lithologic boundaries using the Walsh transform: A case study from the KTB borehole

An automatic method of lithologic boundary detection based on Walsh transform theory is developed and applied to the German Continental Deep Drilling Project (KTB) borehole well log data. Walsh functions are natural choices for describing binary waveforms; hence a Walsh transform-based scheme is most appropriate for detecting lithologic boundaries. The Walsh transform technique involves low-pass filtering augmented by a “Skilled Algorithm” to identify the correct boundary depths. In Walsh domain filtering, the step width of a low-pass image of the original signal is a function of cutoff sequency and also constant over a binary width. This makes the method more suitable to attune and resolve appropriately the minimum thickness of lithologic successions. The application of the method to the best available KTB data yields results that confirm the current geophysical model and also resolve additional finer scale boundaries embedded within larger geological units. The Walsh transform-based detection scheme renders an effective means for resolving and understanding structural inhomogeneity within complex geologic signals, particularly borehole geophysical logs.

Applications of the naturalness preserving transform to image watermarking and data hiding

The paper presents an image watermarking technique using the naturalness preserving transform (NPT). The goal is to produce an invisible and robust watermark that resists signal processing and geometric distortion attacks. Unlike traditional techniques, the proposed algorithm inserts and extracts the watermark via a special orthogonal transform class in between the spatial and frequency representations of the host image data. Two modified forms of the NPT employing the Hartley and discrete cosine transforms, respectively, are presented in order to improve the quality of the transformed image. Enhancements for the quality of the transformed images are achieved via reduction to the artifacts that occur to images transformed via the classical Hadamard-based NPT. The proposed watermark has demonstrated robustness against a variety of image attacks including, JPEG compression, low pass filtering, image cropping, and noise. The marked image must be severely degraded in order to prevent the reconstruction of the watermark.

Efficient, robust, and fast global motion estimation for video coding

In this paper, we propose an efficient, robust, and fast method for the estimation of global motion from image sequences. The method is generic in that it can accommodate various global motion models, from a simple translation to an eight-parameter perspective model. The algorithm is hierarchical and consists of three stages. In the first stage, a low-pass image pyramid is built. Then, an initial translation is estimated with full-pixel precision at the top of the pyramid using a modified n-step search matching. In the third stage, a gradient descent is executed at each level of the pyramid starting from the initial translation at the coarsest level. Due to the coarse initial estimation and the hierarchical implementation, the method is very fast. To increase robustness to outliers, we replace the usual formulation based on a quadratic error criterion with a truncated quadratic function. We have applied the algorithm to various test sequences within an MPEG-4 coding system. From the experimental results we conclude that global motion estimation provides significant performance gains for video material with camera zoom and/or pan. The gains result from a reduced prediction error and a more compact representation of motion. We also conclude that the robust error criterion can introduce additional performance gains without increasing computational complexity.

The effects of spatial frequency overlap on face recognition.

The effects of spatial frequency overlap between pairs of low-pass versus high-pass images on face recognition and matching were examined in 6 experiments. Overlap was defined as the range of spatial frequencies shared by a pair of filtered images. This factor was manipulated by processing image pairs with high-pass/low-pass filter pairs whose 50% cutoff points varied in their separation from one another. The effects of the center frequency of filter pairs were also investigated. In general, performance improved with greater overlap and higher center frequency. In control conditions, the image pairs were processed with identical filters and thus had complete overlap. Even severely filtered low-pass or high-pass images in these conditions produced superior performance. These results suggest that face recognition is more strongly affected by spatial frequency overlap than by the frequency content of the images.

Inverse halftoning using wavelets

This work introduces a new approach to inverse halftoning using nonorthogonal wavelets. The distinct features of this wavelet-based approach are: 1) edge information in the highpass wavelet images of a halftone image is extracted and used to assist inverse halftoning, 2) cross-scale correlations in the multiscale wavelet decomposition are used for removing background halftoning noise while preserving important edges in the wavelet lowpass image, and 3) experiments show that our simple wavelet-based approach outperforms the best results obtained from inverse halftoning methods published in the literature, which are iterative in nature.

A robust single‐shot partial sampling scheme

A method of reducing the amount of data required to reconstruct an image is described. In this scheme, fully sampled low spatial frequency data are acquired up to a given cutoff frequency and above this point, only alternate lines are sampled. Two images are produced, one of low definition and one of high definition but aliased. The proposed algorithm unwraps the aliased data, which are then used to enhance the low pass image, yielding a best estimate of the true image. The reduced sampling technique is shown to afford biological images that are almost indistinguishable from those obtained from a complete data set.

Integration of shape from shading and stereo

Stereo algorithms suffer from the lack of local surface texture due to smoothness of depth constraint, or local miss-matches in disparity estimates. Thus, most stereo methods only provide a coarse depth map which can be associated with a low pass image of the depth map. On the other hand, shape from shading algorithms generally produce better estimates of local surface areas, but some of them have problems with variable albedo and spherical surfaces. Thus, shape from shading methods produce better detailed depth information, and can be associated with the high pass image of the depth map. In order to compute a better depth map, we present a method for integrating the high frequncy information from the shape from shading and the low frequency information from stereo. The proposed algorithm is very simple, takes about 0.7 s for a 128 × 128 image on a Sun SparcStation-1, is non-iterative, and requires very little adjustment of parameters. The results obtained with a variety of synthetic and real images are discussed. The quality of depth obtained by integrating shading and stereo is compared with the ground truth (range image) using height error measure, and improvement ranging from 30 to 50% over stereo, and from 65 to 98% over shading is demonstrated.

Linear filtering of images based on properties of vision

The design of linear image filters based on properties of human visual perception has been shown to require the minimization of criterion functions in both the spatial and frequency domains. We extend this approach to continuous filters of infinite support. For lowpass filters, this leads to the concept of an ideal lowpass image filter that provides a response that is superior perceptually to that of the classical ideal lowpass filter.

Performance of the nonlinear joint transform correlator for images with low-pass characteristics

The correlation performance of the nonlinear joint transform correlator for input images containing a target embedded in input noise is investigated. We focus on a class of images for the target, the input image, and the input noise for which nonlinear joint transform correlators perform well. The target and the input noise have low-pass characteristics; that is, they are not band limited but have most of their energies in the low spatial frequency domain. The analytical results show that for this class of low-pass (but not band-limited) images, the binary joint transform correlator outperforms other types of kth-law nonlinear joint transform correlators with k ≠ 0 when the input-noise bandwidth is less than the target bandwidth. We show that binary nonlinear joint transform correlators perform well when the input noise is signallike; that is, the input noise or the objects to be rejected have energy spectra that are similar to the target energy spectrum. Correlation tests with computer simulations are prese ted. The nonlinear joint transform correlator performance is determined for various degrees of nonlinear transformations and for different input-noise parameters. For the nonlinear joint transform correlator the correlation-signal term and the output-noise term arising from the input noise are determined for various degrees of nonlinearity for the target embedded in input noise.

Combining DCT and subband coding into an intraframe coder

The advent of broadband digital networks will make it possible to transmit high quality digital television. Subband coding is a very interesting coding scheme as it accommodates a transmission quality that can flexibly adapt to the available channel capacity. Subband coding (SBC) and block cosine-transform techniques have already been used together, but only to code the low-pass image. Residual spatial correlation present in higher frequencies bands has only partially been taken care of. This paper presents the results obtained with a simulated coder in which DCT is used in all bands where expected spatial correlation is high enough. The use of a pyramid vector quantizer is proposed. Results are good and compare favourably with other results published in the literature.

Contrast in complex images

The physical contrast of simple images such as sinusoidal gratings or a single patch of light on a uniform background is well defined and agrees with the perceived contrast, but this is not so for complex images. Most definitions assign a single contrast value to the whole image, but perceived contrast may vary greatly across the image. Human contrast sensitivity is a function of spatial frequency; therefore the spatial frequency content of an image should be considered in the definition of contrast. In this paper a definition of local band-limited contrast in images is proposed that assigns a contrast value to every point in the image as a function of the spatial frequency band. For each frequency band, the contrast is defined as the ratio of the bandpass-filtered image at the frequency to the low-pass image filtered to an octave below the same frequency (local luminance mean). This definition raises important implications regarding the perception of contrast in complex images and is helpful in understanding the effects of image-processing algorithms on the perceived contrast. A pyramidal image-contrast structure based on this definition is useful in simulating nonlinear, threshold characteristics of spatial vision in both normal observers and the visually impaired.

New synthetic high coding using edge sharpening

A new synthetic high coding is presented for high data compression. In this scheme, only the lowpass image is transmitted, which is composed of the low-frequency components of an image, through subsampling and adaptive transform coding (ATC). Then it is expanded by interpolation at the receiver. The highpass image is simply synthesised without any overhead information by edge-sharpening the expanded lowpass image. The result of the proposed scheme shows acceptable image quality even at a high compression ratio of about 30:1

The Laplacian Pyramid as a Compact Image Code

We describe a technique for image encoding in which local operators of many scales but identical shape serve as the basis functions. The representation differs from established techniques in that the code elements are localized in spatial frequency as well as in space. Pixel-to-pixel correlations are first removed by subtracting a lowpass filtered copy of the image from the image itself. The result is a net data compression since the difference, or error, image has low variance and entropy, and the low-pass filtered image may represented at reduced sample density. Further data compression is achieved by quantizing the difference image. These steps are then repeated to compress the low-pass image. Iteration of the process at appropriately expanded scales generates a pyramid data structure. The encoding process is equivalent to sampling the image with Laplacian operators of many scales. Thus, the code tends to enhance salient image features. A further advantage of the present code is that it is well suited for many image analysis tasks as well as for image compression. Fast algorithms are described for coding and decoding.