A Real World Multi-UAV Evolutionary Planner for Minimum Time Target Detection

The effects of the degree of structuring and the form of the visual field on target detection and target location were investigated for three different search times. Thirty undergraduate subjects searched for low-visibility targets on a special screen and recorded the target locations on prepared data sheets. The results suggest that both target detection and location are related in a nonmonotonic fashion to the degree of structure and to the form of the field. The structuring continuum was defined as the division of the search area into two or more partitions of equal area. Optima for both measures occurred at the low end of the continuum.Of the two forms investigated, performance with square fields was generally superior to that for circular fields. In answer to a questionnaire after the experiment, a majority of subjects reported a preference for searching the square fields. Both target detection and target location varied directly with amount of search time. However, the functions tended to increase rapidly at first and then level off at the longer search times.

The effect of inherent location uncertainty on the detection of stationary targets was determined in noisy image sequences. Targets were thick and thin projected cylinders mimicking arteries, catheters, and guide wires in medical imaging x-ray fluoroscopy. With the use of an adaptive forced-choice method, detection contrast sensitivity (the inverse of contrast) was measured both with and without marker cues that directed the attention of observers to the target location. With the probability correct clamped at 80%, contrast sensitivity increased an average of 77% when the marker was added to the thin-cylinder target. There was an insignificant effect on the thick cylinder. The large enhancement with the thin cylinder was obtained even though the target was located exactly in the center of a small panel, giving observers the impression that it was well localized. Psychometric functions consisting of d' plotted as a function of the square root of the signal-energy-to-noise-ratio gave a positive x intercept for the case of the thin cylinder without a marker. This x intercept, characteristic of uncertainty in other types of detection experiments, disappeared when the marker was added or when the thick cylinder was used. Inherent location uncertainty was further characterized by using four different markers with varying proximity to the target. Visual detection by human observers increased monotonically as the markers better localized the target. Human performance was modeled as a matched-filter detector with an uncertainty in the placement of the template. The removal of a location cue was modeled by introducing a location uncertainty of approximately equals 0.4 mm on the display device or only 7 microm on the retina, a size on the order of a single photoreceptor field. We conclude that detection is affected by target location uncertainty on the order of cellular dimensions, an observation with important implications for detection mechanisms in humans. In medical imaging, the results argue strongly for inclusion of high-contrast visualization markers on catheters and other interventional devices.

In this paper, we present a proof-of-concept demonstration for the electrification of transportation, aiming to create great opportunities for integrating research and education. Multiple unmanned ground electric vehicles are emulated using Lego Mindstorms EV3 education kits. We implement traffic logic (e.g., path tracking algorithm), voice and gesture control, and collision avoidance functionality into each vehicle. Every vehicle is able to communicate with each other and self-navigate on a 240-sqft platform using local measurement (e.g., touch sensor, color sensor, and infrared sensor). Multiple vehicles are used to simulate the real-world traffic scenarios. Two demo-scale plug-in and wireless charging stations are also built up to mimic the dynamic charging for electric vehicles. Meanwhile, each vehicle can wirelessly communicate with the centralized transportation coordinator to find the best charging spot depending on location, user preference, and real-time pricing. In addition, each vehicle is equipped with an on-board battery monitoring system to measure the real-time battery state (e.g., current, voltage, and temperature) and estimate the battery state-of-charge (SOC). The user-friendly graphical interface is developed in Labview for each component of proof-of-concept demonstration. Most importantly, this demonstration is fully compatible with a variety of general purpose engineering software (e.g., Matlab, and Labview), and creates great opportunities for integrating research and education in Smart Grid and intelligent transportation systems.

Real-time stitching of UAV (Unmanned Aerial Vehicle) sensor images can effectively enhance the sensory ability of the ground station upon the task environment thereof. When UVA sensor images stitched at different time, if the moving object is located at the overlapping part of the adjacent images, the dislocation brought by the object movement may cause phenomenon of the moving object in the final stitched image and accordingly influences the final stitching effect. Therefore, a ghosting elimination method based on target location information was designed in this paper, specifically as follows: firstly, obtaining the location information of the moving object by the target detection method; then, adopting the improved Criminisi algorithm to automatically repair the image from which the dynamic object is eliminated; finally, stitching the images and intensively displaying the extracted target at the corresponding position in a form of three dimensional entity. In this paper, the proposed method was adopted for the real-time stitching test of UVA aerial images, and the experiment result has proved the effectiveness of the proposed method.

Target localization based on energy considerations in distributed sensor networks

Our movements are variable, but the origin of this variability is poorly understood. We examined the sources of variability in human saccadic eye movements. In two experiments, we measured the spatiotemporal variability in saccade trajectories as a function of movement direction and amplitude. One of our new observations is that the variability in movement direction is smaller for purely horizontal and vertical saccades than for saccades in oblique directions. We also found that saccade amplitude, duration, and peak velocity are all correlated with one another. To determine the origin of the observed variability, we estimated the noise in motor commands from the observed spatiotemporal variability, while taking into account the variability resulting from uncertainty in localization of the target. This analysis revealed that uncertainty in target localization is the major source of variability in saccade endpoints, whereas noise in the magnitude of the motor commands explains a slightly smaller fraction. In addition, there is temporal variability such that saccades with a longer than average duration have a smaller than average peak velocity. This noise model has a large generality because it correctly predicts the variability in other data sets, which contain saccades starting from very different initial locations. Because the temporal noise most likely originates in movement planning, and the motor command noise in movement execution, we conclude that uncertainty in sensory signals and noise in movement planning and execution all contribute to the variability in saccade trajectories. These results are important for understanding how the brain controls movement.

The traditional infrared small target detection method is easy to cause false alarms in the complex sea and sky background. In this letter, a fast detection method for infrared small targets in the complex sea and sky background is proposed. Firstly, in order to make full use of the Gaussian characteristics of infrared small targets and effectively distinguish the target from the background clutter, the SODD (second order directional derivative) filter is used to improve the target signal-to-clutter ratio (SCR) and suppress the background clutter; Secondly, in order to improve the real-time performance of the algorithm, we improve the existing MPCM algorithm by reducing the difference dimension between the target area and the local background area; Finally, we use pipeline filtering that can take full advantage of candidate target location, size, and total grayscale 3D information to complete multi-frame target validation and achieve continuous target detection in image sequences. The experimental results show that the proposed algorithm can stably and accurately detect infrared small targets in the infrared image sequence.

On-going research to improve hyperspectral target detection generally focuses on statistical detector performance, reduction of background or environmental contributions to at-sensor radiance, dimension reduction and many other mathematical or physical techniques. These efforts are all aimed at improving target identification in a single scene or data cube. This focus on single scene performance is driven directly by the airborne collection concept of operations (CONOPS) of a single pass per target location. Today's pushbroom and whiskbroom sensors easily achieve single passes and single collects over a target location. If multiple passes are flown for multiple collects on the same location, the time scale for revisit is several minutes. Emerging gimbaled hyperspectral imagers have the capability to collect multiple scans over the same target location in a time scale of seconds. The ability to scan the same location from slightly different collection geometries below the time scale of significant solar and atmospheric change forces us to reexamine the methods for target detection via the fundamental radiance equation. By expanding the radiance equation in the spatial and temporal dimensions, data from multiple hyperspectral images is used simultaneously for determining at-sensor radiance and surface leaving radiance with the ultimate goal of improving target detection. This research reexamines the fundamental radiance equation for multiple scan collection geometries expanding both the spatial and temporal domains. In addition, our assumptions for determining at-sensor radiance are revisited in light of the increased dimensionality. The expanded radiance equation is then applied to data collected by a gimbaled long wave infrared hyperspectral imager. Initial results and future work are discussed.

Locating trapped targets using the signals of wireless devices such as mobile phones carried by targets is increasingly becoming the preferred scheme for disaster rescue. However, the accuracy of radio-localization technology currently adopted in the rescue is not high enough. To solve such problems, researchers proposed the approaches of Warwalking or Wardriving. However, both approaches are all limited by application scenarios. This paper proposes GuideLoc, a highly efficient aerial wireless localization system, which uses the target guiding technology based on region division. GuideLoc allows an unmanned aerial vehicle (UAV) to fly over a target device and provide position coordinates of the UAV as the target location information. During the process of positioning targets, based on the result of regional division, GuideLoc uses a number of fixed antennas to get the received signal strength indicator (RSSI) and angle of arrival (AOA) information of the target and estimates the target location using the information and correspondingly controls the UAV to fly towards the target. Averaging method is also applied for determining coordinates of the target. Experiment results show that GuideLoc can in short time achieve 2.7-meter positioning accuracy in average. Compared with other wireless localization methods using UAVs proposed in the literature, it shortened the flight distance by more than 50%. In addition, energy efficiency is also improved significantly.

Space time coding plays quite an important role in the design of multi-input multi-output (MIMO) radar. It has been shown that the correlation characteristics of waveform set have a great impact on the performance of target detection and localisation for MIMO radar. In this study, the target detection and localisation performance of co-located MIMO radar is analysed theoretically and experimentally, when correlated (not ideally orthogonal or fully coherent) waveforms are transmitted. The relationship between transmit–receive beam pattern improvement and correlation coefficients of transmitted waveforms is investigated. The signal to interference and noise ratio varying with correlation coefficients in range-Doppler matched processing is defined and analysed. Furthermore, Cramer–Rao bound for target direction of arrival estimation and target detection probability in the Neyman–Pearson criteria are derived when transmitting correlated waveforms. It is shown that the MIMO radar detection and localisation performance is much dependent with the correlation level of transmitted waveforms, which can be quite useful for MIMO radar orthogonal waveform design. Numerical simulation validates the theoretical analysis.

The detection of small infrared targets lacking texture and shape information in the presence of complex background clutter is a challenge that has attracted considerable research attention in recent years. Typical deep learning-based target detection methods are designed with deeper network structures, which may lose targets in the deeper layers and cannot directly be used for small infrared target detection. Therefore, we designed the attention fusion feature pyramid network (AFFPN) specifically for small infrared target detection. Specifically, it consists of feature extraction and feature fusion modules. In the feature extraction stage, the global contextual prior information of small targets is first considered in the deep layer of the network using the atrous spatial pyramid pooling module. Subsequently, the spatial location and semantic information features of small infrared targets in the shallow and deep layers are adaptively enhanced by the designed attention fusion module to improve the feature representation capability of the network for targets. Finally, high-performance detection is achieved through the multilayer feature fusion mechanism. Moreover, we performed a comprehensive ablation study to evaluate the effectiveness of each component. The results demonstrate that the proposed method performs better than state-of-the-art methods on a publicly available dataset. Furthermore, AFFPN was deployed on an NVIDIA Jetson AGX Xavier development board and achieved real-time target detection, further advancing practical research and applications in the field of unmanned aerial vehicle infrared search and tracking.

BackgroundWe recently showed that the empirical use of a combination of 2 moving average (MA) statistical process control (SPC) charts was highly sensitive and specific for detecting potentially important increases in surgical site infection (SSI) rates. We performed this follow-up study to examine the performance of these same SPC charts when applied to known SSI outbreaks.MethodsWe retrospectively applied 2 MA SPC charts to all 30 SSI outbreaks investigated from 2007 to 2015 in a network of over 50 community hospitals. These outbreaks were detected via routine SSI surveillance activities that occurred in the network. We reviewed prior outbreak investigation documentation to determine the estimated time of outbreak onset and time of traditional surveillance outbreak detection. The first SPC chart utilized procedure-specific, composite SSI data from the hospital network for its baseline; the baseline for the second chart was calculated from SSI data from the outbreak hospital undergoing analysis. Both charts used rolling baseline windows but varied in baseline window size, rolling baseline lag, and MA window size. SPC chart outbreak detection occurred when either chart had a data point above the upper control limit of 1 standard deviation. Time of SPC detection was compared with both time of outbreak onset and time of traditional surveillance detection.ResultsWith the dual chart approach, SPC detected all 30 outbreaks, including detection of 25 outbreaks (83%) prior to their estimated onset (Figure 1). SPC detection occurred a median of 16 months (interquartile range, 12–21 months) prior to the date of traditional outbreak detection, which never occurred prior to outbreak onset. Both individual SPC charts exhibited at least 90% sensitivity in outbreak detection, but the dual chart approach showed superior sensitivity and speed of detection (Figure 2).ConclusionA strategy that employed optimized, dual MA SPC charts retrospectively detected all SSI outbreaks that occurred over 9 years in a network of community hospitals. SPC outbreak detection occurred earlier than traditional surveillance detection. These optimized SPC charts merit prospective study to evaluate their ability to promote early detection of SSI clusters in real-world scenarios.DisclosuresAll Authors: No reported Disclosures.

JDECMC: Improving JDE based multi-object tracking with Camera Motion Compensation

With the increasing complexity of marine missions, it is necessary to resolve the missions through coordination of multiple unmanned underwater vehicle. This paper combines the virtual leader strategy with passivity-based techniques to solve the problem of coordination path following of multiple unmanned underwater vehicle. Meanwhile, the coordinated path following with a desired spatial formation is achieved through defining the formation reference point. The consensus of formation reference point is realized by using the synchronization controller based on passivity. Finally, the effectiveness of the proposed algorithm is verified through simulation.

Abstract. Target detection is a vital field within computer vision, playing an essential role in applications. This paper investigates the advancements and efficiencies of contemporary target detection methodologies, focusing on deep learning frameworks. Through a systematic review and evaluation of these models' architectures and performances using the Common Objects in Context (COCO) dataset, the study highlights their operational effectiveness and practical implications in real-world scenarios. A detailed comparative analysis is conducted, assessing the models based on mean Average Precision (mAP) and input dimensions to determine their suitability across various detection tasks. You Only Look Once version 3 (YOLOv3), in particular, is recognized for its ability to combine high-speed detection with significant accuracy, effectively addressing real-time processing challenges. The results confirm that YOLOv3, despite its smaller input size, performs comparably to more complex systems, demonstrating notable enhancements in design efficiency and processing speed. This research underscores the potential for future optimizations in model architectures to bridge the gap between high-speed and high-accuracy detection tasks, potentially transforming real-time applications across multiple sectors. The practical significance of this work lies in its ability to guide future developments in object detection, benefiting both academic research and industrial applications.