It is important to understand the factors underlying grade crossing crashes, and to examine potential solutions. We have installed a camera in front of a locomotive to examine grade crossing accidents (or near accidents). We present a computer vision system that automatically extracts possible near accidents scenes by detecting the activity of vehicles crossing in front of the train after the signals are ignited. We presented a fast algorithm to detect moving objects that is recorded by a moving camera with minimal computation. The moving object is detected by 1) estimating ego-motion of the camera and 2) detecting and tracking feature points whose motion is inconsistent with the camera motion. We introduce a pseudo-realtime ego-motion (camera motion) estimation method with a robust optimization algorithm. We present experiments on ego-motion estimation and moving object detection. Our algorithm works in pseudo-realtime and we expect that our algorithm can be applied to realtime applications, such as collision warning, in the near future with the development of hardware technology.
In real-world signaling situations, rapid detection of a signal can be vitally important. One such situation involves a following vehicle (FV) that is too close to a lead vehicle (LV). Suppose that the LV could detect the problem. How should the FV be signaled? We studied visual warning signals typical of those used on large LV's, like buses: in this case, an eight-segment 8 CM × 150 CM horizontal array. | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | The endpoint measure was time to react to light onset in the array (RT). The ‘standard’ signal involved simultaneous ignition of all eight segments at t=0. The ‘test’ signal delayed the ignition of some segments (first 4,5 @ t=0; then 3,6; then 2,7; then 1,8, each step in the sequence delayed by 50MSEC) producing an apparently expanding pattern. Observers included 6 pre-presbyopic, normal, corrected adults plus 3 corrected adults over 60. Conditions that were varied during testing included adaptation level, array intensity and array size (19 DEG or 1.9 DEG). 25 repetitions of each signal were tested for each condition. Paradoxically, faster RT's occurred with piecemeal ignition. Median RT's were sometimes faster by as much as 50 MSEC and once, by 100 MSEC for individuals, compared to RT's for simultaneous ignition. The difference was due mainly to increased numbers of late RT's for the ‘standard’. In nearly every condition, the ‘test’ signal was seen significantly more quickly than the simultaneous signal. The exception involved a ‘wig-wag’ signal in which we turned on 1,2,5, &6 then, later, 3,4,7,&8. The paradox can be explained if one presumes that the standard signal does not stimulate M-neurons (or, some might suggest, ‘looming’ detectors) as well as does the piecewise-delayed ‘test’ signal. To determine whether a real-world LV-FV situation can benefit from this knowledge, a delayed test-signal will soon be deployed on a transit bus LV equipped with a radar-based FV detector.
Support data for UMMZ Miscellaneous Publication No. 203. An Excel spreadsheet used to generate the maps in study. Enabling reader to generate digitized maps.
In Fleming's original paper on penicillin, he described the lysis of staphylococci by this agent. Since then the lytic process and ancillary morphologic changes in the bacteria concerned have been studied by several investigators. We wish to report here the effect of penicillin on the lytic action of bacteriophage. The phage employed in our experiments is the “K” race; it and the homologous strain of Staphylococcus aureus have been used in studies described in a series of papers appearing since 1929. Stock cultures were grown in Roux flasks for 18 hours at 36°C. In order to have a substrate of actively growing cells, 1 × 108 staphylococci/ml were suspended in 100 ml of broth contained in a 500 ml flask; aeration was provided by placing the flask in a shaker operating in a water bath set at 36°C. When growth had progressed to a level of 1 × 109 bacteria/ml, the cell suspension was removed from the shaker and used for the experiments described below. To determine [bacteria] when values were >5 × 108 staphylococci/ml, the direct microscopic count method was employed. For lytic curves, visual comparison was made with standards covering a range from 2 × 107 to 5 × 108 staphylococci/ml in formalinized broth. The fluid medium throughout was tryptose phosphate broth. Phage determinations were carried out by Gratia's method of counting plaques and the values noted below are in plaques/ml. A 10 ml mixture containing 5 × 108 staphylococci/ml, 5 × 108 phage units/ml and 10 units of sodium penicillin/ml was placed in a test tube and shaken at 36°C. Turbidity readings were made every 0.2 hour. This was also done with a suspension from which the phage was omitted and with one containing no penicillin.
Grosser and Spafford (in this journal, 1989) have advanced an hypothesis and presented measurements which they believe support the idea of an excess of cones in the peripheral retinae of dyslexics. This note points out that their hypothesis is based on the erroneous assumption that normals have no peripheral cones. Further, their data can be explained by at least two alternative, though uninteresting, methodological hypotheses, that uncontrolled eye movements or experimenter suggestion (or both) could have produced their results. Finally, the requisite methods for assessing color vision, and the cones, were not met in the study.
This project centers on the development and testing of an integrated display of warning signals for the intelligent transportation systems (ITS) vehicles driver. Plans for the highway of the future predict as many as 12 new warning signals whose purpose is to alert the driver to incipient difficulties related to vehicle operation, proximity of other vehicles, driving conditions, and vehicle position. In this study, the optimal configuration of these signals at the driver's eye is explored. The aim of the study is to maximize detectability and identifiability of such signals while minimizing the time to react. Accumulated research in the human factors and vision literature points to a number of constraints and desirable properties that a warning signal display must satisfy. Using this work, plus known characteristics of the human visual system, a set of specifications for the warning signal display is developed. Specifications include temporal, chromatic, and spatial characteristics of the visual stimuli to be used as warning signals. A proprietary process is incorporated in the design that is thought to optimize both signal detectability and salience by supplying visual warning signal optimized for processing by the fastest and most sensitive nervous pathway in the human visual system. The project will rigorously test human performance with a display that satisfied these specifications and will provide the ITS community with a description of design constraints, an example of a design developed within these constraints, benchmark performance tests of it, and a means of evaluating future prototypes developed for commercial application.
With rear-end collisions being the most common crash experienced by transit buses and with the country's estimated 45,000 transit buses suffering a crash about once a year, this research focused on a device to reduce rear-end collisions. The subject of this report is a light bar, comprised of eight separate units, that sits on the rear of the bus at about eye-level to a driver and would be intended to warn the driver of the need to stop. It is intended to ultimately be used in conjunction with a detection system on the rear of the bus that would cause the warning to go off when it sensed an impending collision. Replacing incandescent bulbs with LEDs reduces the time needed for the bar to light significantly. The next task was to determine the pattern for igniting the eight units that drivers would perceive most rapidly. Tests indicated that turning on the units sequentially enabled observers to see them significantly sooner, enough to provide an extra 4.4 feet of stopping room on average for a car traveling at 30 mph.
