SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Computer program corrects various photometic, geometric and frequency response distortions in pictures. The program converts pictures to a number of elements, with each elements optical density quantized to a numerical value. The translated picture is recorded on magnetic tape in digital form for subsequent processing and enhancement by computer.
Photogeologic and computer enhancement of geological imagery of Colorado plateau region of Arizona
A computer program was developed to generate random numbers on the Sigma V computer. A subroutine gave a random real number from a uniform distribution on the interval (0,1). A normal distribution with mean zero and variance 1 was also obtained.
Image-analyzing pointing systems aimed to high precision. Star-tracker program, STRACKER, developed to solve algorithm-design problems for area-array tracking and pointing systems operating at accuracies of 0.001 to 0.01 picture element. Includes auxiliary programs for reformatting point-spread data from commercial ACCOSV lens-design program. Other optical-analysis program data reformatted by use of utility routines included in package. Written in FORTRAN 77.
Robust vision capability for intelligent control systems has been an elusive goal in image processing. The computationally intensive techniques a necessary for conventional image processing make real-time applications, such as object tracking and collision avoidance difficult. In order to endow an intelligent control system with the needed vision robustness, an adequate image enhancement subsystem capable of compensating for the wide variety of real-world degradations, must exist between the image capturing and the object recognition subsystems. This enhancement stage must be adaptive and must operate with consistency in the presence of both statistical and shape-based noise. To deal with this problem, we have developed an innovative algebraic approach which provides a sound mathematical framework for image representation and manipulation. Our image model provides a natural platform from which to pursue dynamic scene analysis, and its incorporation into a vision system would serve as the front-end to an intelligent control system. We have developed a unique polynomial representation of gray level imagery and applied this representation to develop polynomial operators on complex gray level scenes. This approach is highly advantageous since polynomials can be manipulated very easily, and are readily understood, thus providing a very convenient environment for image processing. Our model presents a highly structured and compact algebraic representation of grey-level images which can be viewed as fuzzy sets.
An input image is enhanced to include spatial frequency components having frequencies higher than those in an input image. To this end, an edge map is generated from the input image using a high band pass filtering technique. An enhancing map is subsequently generated from the edge map, with the enhanced map having spatial frequencies exceeding an initial maximum spatial frequency of the input image. The enhanced map is generated by applying a non-linear operator to the edge map in a manner which preserves the phase transitions of the edges of the input image. The enhanced map is added to the input image to achieve a resulting image having spatial frequencies greater than those in the input image. Simplicity of computations and ease of implementation allow for image sharpening after enlargement and for real-time applications such as videophones, advanced definition television, zooming, and restoration of old motion pictures.
The Space Shuttle Orbiter windows were damaged both by micrometeor impacts and by handling, and required careful inspection before they could be reused. The launch commit criteria required that no defect be deeper than a critical depth. The shuttle program used a refocus microscope to perform a quick pass/fail determination, and then followed up with mold impressions to better quantify any defect. However, the refocus microscope is slow and tedious to use due to its limited field of view, only focusing on one small area of glass at a time. Additionally, the unit is bulky and unable to be used in areas with tight access, such as defects near the window frame or on the glass inside the Orbiter due to interference with the dashboard. The surface inspection tool is a low-profile handheld instrument that provides two digital video images on a computer for monitoring surface defects. The first image is a wide-angle view to assist the user in locating defects. The second provides an enlarged view of a defect centered in the window of the first image. The focus is adjustable for each of the images. However, the enlarged view was designed to have a focal plane with a short depth. This allows the user to get a feel for the depth of different parts of the defect under inspection as the focus control is varied. A light source is also provided to illuminate the defect, precluding the need for separate lighting tools. The software provides many controls to adjust image quality, along with the ability to zoom digitally the images and to capture and store them for later processing.
The effect was examined of aircraft position and attitude, of measurement errors, and of computational approximations on the size, shape, and position of a perspective radar image of an airport runway as might be displayed by an independent landing monitor in transport aircraft. The effect on runway image geometry was examined for different aircraft attitudes and different aircraft positions relative to a standard three degree glide slope. Measurement errors investigated were errors in radar azimuth angle and range, and errors in those aircraft parameters supplied to the radar for use in converting the radar image into a perspective format (namely pitch, roll, and altitude). Also investigated were the effects of using certain mathematical approximations, such as small angle, in the coordinate transformation which converts the image to a perspective format.
Image processing technique quantifies human atherosclerosis by computer analysis of arterial angiograms. X-ray film images are scanned and digitized, arterial shadow is tracked, and several quantitative measures of lumen irregularity are computed. In other tests, excellent agreement was found between computer evaluation of femoral angiograms on living subjects and evaluation by teams of trained angiographers.
ENHANCETOOL computer program has capabilities for interactive enhancement of digital images. Includes particularly useful combination of algorithms not existing in single interactive program. Software package also provides means through which additional image-enhancement algorithms easily integrated and made available to user. Written in C Language.
A computer program automatically builds large, full-resolution mosaics of multispectral images of Earth landmasses from images acquired by Landsat 7, complete with matching of colors and blending between adjacent scenes. While the code has been used extensively for Landsat, it could also be used for other data sources. A single mosaic of as many as 8,000 scenes, represented by more than 5 terabytes of data and the largest set produced in this work, demonstrated what the code could do to provide global coverage. The program first statistically analyzes input images to determine areas of coverage and data-value distributions. It then transforms the input images from their original universal transverse Mercator coordinates to other geographical coordinates, with scaling. It applies a first-order polynomial brightness correction to each band in each scene. It uses a data-mask image for selecting data and blending of input scenes. Under control by a user, the program can be made to operate on small parts of the output image space, with check-point and restart capabilities. The program runs on SGI IRIX computers. It is capable of parallel processing using shared-memory code, large memories, and tens of central processing units. It can retrieve input data and store output data at locations remote from the processors on which it is executed.
Review of some of the computer-aided digital image processing techniques recently developed. Special attention is given to mapping and mosaicking techniques and to preliminary developments in range determination from stereo image pairs. The discussed image processing utilization areas include space, biomedical, and robotic applications.
Advancements were made in automating microscopy-based material characterization, particularly in studying irradiation effects on additively manufactured (AM) materials using machine learning (ML) and computer vision (CV). These automation efforts address the challenges of analyzing complex microstructures, accelerating the detection of irradiation-induced defects. Two CV models were developed at Argonne National Laboratory (ANL) to enhance transmission electron microscopy (TEM) analysis of irradiated AM 316 stainless steel. The first model focused on the detection of irradiation-induced dislocation loops, which contribute to material hardening and embrittlement. These loops, categorized as faulted or perfect, were automatically detected and classified using a Mask R-CNN model trained on TEM images from both in-situ and ex-situ ion irradiation experiments. The model achieved high accuracy, with precision, recall, and F1 scores of 0.839, 0.734, and 0.776, respectively, demonstrating its effectiveness in analyzing dislocation loops in irradiated AM materials. The second CV model was developed to analyze the size and wall thickness of dislocation cells in laser powder bed fusion (LPBF) 316 stainless steel. Using a U-Net++ architecture with EfficientNet as the encoder, the model was trained on TEM images to segment and measure cell size and wall thickness.
A digital image compression preprocessor for use in a discrete cosine transform-based digital image compression device is provided. The preprocessor includes a gathering mechanism for determining discrete cosine transform statistics from input digital image data. A computing mechanism is operatively coupled to the gathering mechanism to calculate a image distortion array and a rate of image compression array based upon the discrete cosine transform statistics for each possible quantization value. A dynamic programming mechanism is operatively coupled to the computing mechanism to optimize the rate of image compression array against the image distortion array such that a rate-distortion-optimal quantization table is derived. In addition, a discrete cosine transform-based digital image compression device and a discrete cosine transform-based digital image compression and decompression system are provided. Also, a method for generating a rate-distortion-optimal quantization table, using discrete cosine transform-based digital image compression, and operating a discrete cosine transform-based digital image compression and decompression system are provided.
A comprehensive study has been undertaken to determine the extent to which ERTS-1 data could be used to detect, identify (classify), locate and measure features of applications interest in the disciplines of Agriculture and Forestry. The study areas included: six counties in five states in which were located examples of the most important crops and practices of American agriculture; and a portion of the Sam Houston National Forest, a typical Gulf coastal plain pine forest. The investigation utilized conventional image interpretation and computer-aided (spectral pattern recognition) analysis using both image products and computer compatible tapes. The emphasis was generally upon the computer-aided techniques. It was concluded that ERTS-1 data can be used to detect, identify, locate and measure a wide array of features of interest in agriculture and forestry.
Six channels of multispectral middle infrared (8 to 14 micron) aircraft scanner data were acquired over the East Tintic mining district, Utah. This area has high relief and moderate vegetation and consists mainly of Tertiary silicic igneous rocks and Paleozoic quartzite and carbonate rocks that have been locally hydrothermally altered. These digital-image data were computer processed to create a color-composite image based on principal component transformations. Color differences in this image are related to the spectral differences in the surface material and allow discrimination of several rock types, depending primarily on their silica content. When combined with a visible and near infrared color-composite image from a previous flight, with limited field checking, it is possible to discriminate quartzite, carbonate rocks, quartz latitic and quartz monzonitic rocks, latitic and monzonitic rocks, silicified altered rocks, argillized altered rocks, and vegetation.
We implement a quantum generalization of a neural network on trapped-ion and IBM superconducting quantum computers to classify MNIST images, a common benchmark in computer vision. The network feedforward involves qubit rotations whose angles depend on the results of measurements in the previous layer. The network is trained via simulation, but inference is performed experimentally on quantum hardware. The classical-to-quantum correspondence is controlled by an interpolation parameter, $a$, which is zero in the classical limit. Increasing $a$ introduces quantum uncertainty into the measurements, which is shown to improve network performance at moderate values of the interpolation parameter. We then focus on particular images that fail to be classified by a classical neural network but are detected correctly in the quantum network. For such borderline cases, we observe strong deviations from the simulated behavior. We attribute this to physical noise, which causes the output to fluctuate between nearby minima of the classification energy landscape. Such strong sensitivity to physical noise is absent for clear images. We further benchmark physical noise by inserting additional single-qubit and two-qubit gate pairs into the neural network circuits. Our work provides a springboard toward more complex quantum neural networks on current devices: while the approach is rooted in standard classical machine learning, scaling up such networks may prove classically non-simulable and could offer a route to near-term quantum advantage.
Entire process under control of computer. Computer-controlled image-acquisition system tracks object, such as water bubble, as it moves in drop tube. Ultimately, such tracking system used to observe fusion-pellet formation in drop furnace.