In this paper, we describe a speed estimation method for individual vehicles using a monocular camera. The system includes the following: (1) object detection, which detects an object of interest based on a combination of motion detection and object classification and initializes tracking of the object if detected, (2) object tracking, which tracks the object over time based on template matching and reports its frame-to-frame displacement in pixels, (3) speed estimation, which estimates vehicle speed by converting pixel displacements to distances traveled along the road, (4) object height estimation, which estimates the distance from tracked point(s) of the object to the road plane, and (5) speed estimation with height-correction, which adjusts previously estimated vehicle speed based on estimated object and camera heights. We demonstrate the effectiveness of our algorithm on 30/60 fps videos of 300 vehicles travelling at speeds ranging from 30 to 60 mph. The 95-percentile speed estimation error was within ±3% when compared to a lidar-based reference instrument. Key contributions of our method include (1) tracking a specific set of feature points of a vehicle to ensure a consistent measure of speed, (2) a high accuracy camera calibration/characterization method, which does not interrupt regular traffic of the site, and (3) a license plate and camera height estimation method for improving accuracy of individual vehicle speed estimation. Additionally, we examine the impact of spatial resolution on accuracy of speed estimation and utilize that knowledge to improve computation efficiency. We also improve accuracy and efficiency of tracking over standard methods via dynamic update of templates and predictive local search.
There is a worldwide effort to apply 21st century intelligence to evolving our transportation networks. The goals of smart transportation networks are quite noble and manifold, including safety, efficiency, law enforcement, energy conservation, and emission reduction. Computer vision is playing a key role in this transportation evolution. Video imaging scientists are providing intelligent sensing and processing technologies for a wide variety of applications and services. There are many interesting technical challenges including imaging under a variety of environmental and illumination conditions, data overload, recognition and tracking of objects at high speed, distributed network sensing and processing, energy sources, as well as legal concerns. This paper presents a survey of computer vision techniques related to three key problems in the transportation domain: safety, efficiency, and security and law enforcement. A broad review of the literature is complemented by detailed treatment of a few selected algorithms and systems that the authors believe represent the state-of-the-art.
Urban parking management is receiving significant attention due to its potential to reduce traffic congestion, fuel consumption, and emissions. Real-time parking occupancy detection is a critical component of on-street parking management systems, where occupancy information is relayed to drivers via smart phone apps, radio, Internet, on-road signs, or global positioning system auxiliary signals. Video-based parking occupancy detection systems can provide a cost-effective solution to the sensing task while providing additional functionality for traffic law enforcement and surveillance. We present a video-based on-street parking occupancy detection system that can operate in real time. Our system accounts for the inherent challenges that exist in on-street parking settings, including illumination changes, rain, shadows, occlusions, and camera motion. Our method utilizes several components from video processing and computer vision for motion detection, background subtraction, and vehicle detection. We also present three traffic law enforcement applications: parking angle violation detection, parking boundary violation detection, and exclusion zone violation detection, which can be integrated into the parking occupancy cameras as a value-added option. Our experimental results show that the proposed parking occupancy detection method performs in real-time at 5 frames/s and achieves better than 90% detection accuracy across several days of videos captured in a busy street block under various weather conditions such as sunny, cloudy, and rainy, among others.
Video cameras are widely deployed along city streets, interstate highways, traffic lights, stop signs and toll booths by entities that perform traffic monitoring and law enforcement. The videos captured by these cameras are typically compressed and stored in large databases. Performing a rapid search for a specific vehicle within a large database of compressed videos is often required and can be a time-critical life or death situation. In this paper, we propose video compression and decompression algorithms that enable fast and efficient vehicle or, more generally, event searches in large video databases. The proposed algorithm selects reference frames (i.e., I-frames) based on a vehicle having been detected at a specified position within the scene being monitored while compressing a video sequence. A search for a specific vehicle in the compressed video stream is performed across the reference frames only, which does not require decompression of the full video sequence as in traditional search algorithms. Our experimental results on videos captured in a local road show that the proposed algorithm significantly reduces the search space (thus reducing time and computational resources) in vehicle search tasks within compressed video streams, particularly those captured in light traffic volume conditions.
There is a world-wide effort to apply 21st century intelligence to evolving our transportation networks. The goals of
smart transportation networks are quite noble and manifold, including safety, efficiency, law enforcement, energy
conservation, and emission reduction. Computer vision is playing a key role in this transportation evolution. Video
imaging scientists are providing intelligent sensing and processing technologies for a wide variety of applications and
services. There are many interesting technical challenges including imaging under a variety of environmental and
illumination conditions, data overload, recognition and tracking of objects at high speed, distributed network sensing and
processing, energy sources, as well as legal concerns. This conference presentation and publication is brief introduction
to the field, and will be followed by an in-depth journal paper that provides more details on the imaging systems and
It is of great value to be able to track image quality of a printing system and detect changes before/when it occurs. To do
that effectively, image quality data need to be constantly gathered and processed. A common approach is to print and
measure test-patterns over-time at a pre-determined schedule and then analyze the measured image quality data to
discover/detect changes. But due to the presence of other printer noise, such as page-to-page instability, mottle etc., it is
likely that the measured image quality data for a given image quality attribute of interest (e.g. streaks) at a given time is
governed by a statistical model rather than a deterministic one. This imposes difficulty for methods intended to detect
image quality changes reliably unless sufficient data of test samples are collected. However, these test samples are non-value-
add to the customers and should be minimized. An alternative is to directly measure and assess the image quality
attributes of interest from customer pages and post-processing them for detecting changes. In addition to the difficulty
caused by sources of other printer noise, variable image contents from customer pages further impose challenges in the
change detection. This paper addresses these issues and presents a feasible solution in which change points are detected
by statistical model-ranking.
In many color measurement applications, such as those for color calibration and profiling, "patch code" has been used
successfully for job identification and automation to reduce operator errors. A patch code is similar to a barcode, but is
intended primarily for use in measurement devices that cannot read barcodes due to limited spatial resolution, such as
spectrophotometers. There is an inherent tradeoff between decoding robustness and the number of code levels available
for encoding. Previous methods have attempted to address this tradeoff, but those solutions have been sub-optimal. In
this paper, we propose a method to design optimal patch codes via device characterization. The tradeoff between
decoding robustness and the number of available code levels is optimized in terms of printing and measurement efforts,
and decoding robustness against noises from the printing and measurement devices. Effort is drastically reduced relative
to previous methods because print-and-measure is minimized through modeling and the use of existing printer profiles.
Decoding robustness is improved by distributing the code levels in CIE Lab space rather than in CMYK space.
Gamut mapping is a critical process in the output color management of a printer. In systems consisting of two or more
marking engines, these engines could in general have different output color gamuts. In special cases these gamuts can be
drastically different. When applying gamut mapping to a multi-engine printing system, which gamut to use is not
obvious. In the default case, i.e., if the multi-engine system is treated as a single engine with a single gamut, poor image
quality could result. In this paper, we propose gamut mapping strategies for multiple engine printing, which depend on
the relative importance of "color rendition" and "page-to-page color consistency" within a given document. In
particular, we propose a content-based gamut mapping, wherein the trade-offs between color rendition and page-to-page
color consistency can be automatically and dynamically made by segmenting the document and applying gamut mapping
according to the needs of the segmented components.
Mottle is a common defect in printing. Mottle evaluation is crucial in image quality assessment and system
diagnosis. In this paper, we present a new automatic mottle estimation method which improves the existing technologies
in two aspects. First, a modified mottle noise frequency range is proposed, which further separates the banding and
streak spectra from mottle spectrum. Second, a robust estimation algorithm is introduced. It is less sensitive to the
outliers that may appear in the measurement. These outliers include other defects within the mottle frequency range,
such as spots, or defects outside of mottle frequency range, but are strong enough that can not be completely eliminated
by normal spatial filtering.
The method of paired comparisons is often used in image quality evaluations. Psychometric scale values for quality
judgments are modeled using Thurstone's Law of Comparative Judgment in which distance in a psychometric scale
space is a function of the probability of preference. The transformation from psychometric space to probability is a
cumulative probability distribution.
The major drawback of a complete paired comparison experiment is that every treatment is compared to every other,
thus the number of comparisons grows quadratically. We ameliorate this difficulty by performing paired
comparisons in two stages, by precisely estimating anchors in the psychometric scale space which are spaced apart
to cover the range of scale values and comparing treatments against those anchors.
In this model, we employ a generalized linear model where the regression equation has a constant offset vector
determined by the anchors. The result of this formulation is a straightforward statistical model easily analyzed using
any modern statistics package. This enables model fitting and diagnostics.
This method was applied to overall preference evaluations of color pictorial hardcopy images. The results were
found to be compatible with complete paired comparison experiments, but with significantly less effort.
We describe a method for automatically detecting streaks in printed images using adaptive window-based image projections and mutual information. The proposed approach accepts a scanned image enclosing the defect and computes the projections across the entire image at different window sizes. The resulting traces collected from the projections are analyzed with a peak detection algorithm and subsequently correlated using normalized mutual information to pinpoint the location and width of streak(s). Finally, for a given peak, the window size is changed adaptively to identify and locate the intensity and length of the corresponding streak(s) while maximizing the signal-to-noise ratio. Results on synthetic and real-life images are provided to demonstrate the effectiveness of our proposed technique.
We describe a method for automatically classifying image-quality defects on printed documents. The proposed approach accepts a scanned image where the defect has been localized a priori and performs several appropriate image processing steps to reveal the region of interest. A mask is then created from the exposed region to identify bright outliers. Morphological reconstruction techniques are then applied to emphasize relevant local attributes. The classification of the defects is accomplished via a customized tree classifier that utilizes size or shape attributes at corresponding nodes to yield appropriate binary decisions. Applications of this process include automated/assisted diagnosis and repair of printers/copiers in the field in a timely fashion. The proposed technique was tested on a database of 276 images of synthetic and real-life defects with 94.95% accuracy.
It is well-known that many sub-attributes of line quality contribute to the perception of the overall line quality. But the relative importance of these sub-attributes is not clear, nor is there a method available for combining them into one representative number for overall line quality. To address these issues, we have designed and conducted a series of psychophysical experiments, which explore the shape of the human visual transfer functions (VTF) relevant to the perception of three selected sub-attributes: lumpiness, waviness and raggedness. We found that human sensitivity to these sub-attributes can be represented by VTF’s of the same shape but with relative perception weighting factors of 6:4:3 respectively. Based on this, we have proposed an approach to assess overall line quality. In our method, we first pre-process the line image acquired and extract certain profiles relevant to line quality measurement. A set of corresponding VTF’s is then applied to these profiles to calculate the various sub-attributes. Finally, overall line quality is determined by the weighted combination of these individual sub-attributes. These preference weights (1:1:3 for lumpiness, waviness and raggedness respectively) are different from the perception weights mentioned earlier. Our preliminary results show that this measurement correlates well with human perception of overall line quality, for the sub-attributes studied.