Localization of a moving target using a fly eye sensor.

An optical sensor based on the visual system of the common housefly (Musca domestica) has been developed by the researchers at the Wyoming Image and Signal Processing Research (WISPR) Laboratory. This optical sensor shows promising peformance for detection of an edge in real-time, with minimal processing overhead. An application of this sensor might be the deflection measurement of a moving target, such as the wing of an aircraft under test. In this paper, we have investiaged the possibility of localizing an edge, and hence the target, using the fly eye sensor, which is an important part of the deflection measurement process. A simulation program has been developed for this purpose, which simulates the sensor output, for a moving object on the target plane at a specified distance from the sensor. After successfully simulating the sensor output, it has been found that the location of an edge changes with the change in sensor to target distance, and their relative orientation. It also changes for the limited movement of the target, withing the field of view (FOV) of the sensor. On the contrary, the target-background contrast scenario, and shifting of the target normal to the target movement direction, do not affect the edge localization process. This paper shows and edge can be precisely detected for a target a aspecified distance from the sensor with proper sensor-target orientation, when the target moves all the way across the FOV of the sensor.

Effect of sensor-target-background distance on target tracking using a fly eye sensor.

A multi-aperture optical sensor, known as a fly eye sensor, has been developed at the Wyoming Image and Signal Processing Research (WISPR) Laboratory based on the visual system of the common housefly Musca domestical. This biomimetic sensor shows promising edge detection capability, in varying contrast scenarios, with minimal processing overhead. Use of this sensor for fast motion detection, and object tracking is appealing, but optimizing the use of such a sensor requires detailed study. This paper analyzes the effect of placing the background at various distances greater than the target, and provides visualization of these example scenarios. A computer simulationof the sensor using MATLAB demonstrates that the placdementof a target closer to the sensor, and further from the background, affects the sensor response. If not properly considered, this may introduce ambiguities and degrade the performance of a tracking system based ont he flye eye sensor that requires precise location of the target in front of the sensor. This paper shows how a peroperly designed low-pass filter can greatly mitigate this effect with only a small degradation of the relative response magnitude at different distances from the sensor.

Results of target tracking with a Musca domestica inspired sensor.

As a follow-up to previous work done at the University of Wyoming (and presented at a previous Rocky Mountain Bioengineering Symposium), this paper discusses the results of using a Musca domestica based sensor platform as a target tracking mechanism for the measurement of wing deflection of fixed-wing aircraft. The testing of the sensor hardware and accompanying software is described, and the results are analyzed. Work remains to be done to improve robustness and adaptability, but given specific operating conditions, the sensor is a viable alternative to other technologies, and provides results with improved efficiency, speed, and computational load.

Changes, adaptations, and applications of a bio inspired machine vision sensor - biomed 2013.

The Musca domestica (common housefly) biomimetic sensor project is an ongoing endeavor at the University of Wyoming. The project has developed a sensor platform based on the rudimentary function of a fly’s compound eye, including the inherent hyperacuity (high sensitivity to motion) present in the insect’s vision. The design portion of the project is nearing completion, and application driven characterization can now begin. NASA is funding research into utilizing the sensor for efficient, fast, and inexpensive target tracking applications. This stage of the project is very preliminary, but is advancing and offers unique challenges and opportunities. his paper provides background information on the sensor design being considered, discusses the advances in the most current sensor platform, and offers an application currently being investigated.

Signal artifacts in a light-adapted, Musca domestica-based sensor system - biomed 2011.

The Musca domestica (common housefly) sensor project is an ongoing endeavor at the University of Wyoming. The project seeks to develop a sensor based on the rudimentary function of a fly’s eye, including the inherent hyperacuity (high sensitivity to motion) present in the insect’s vision. During the summer of 2010, several characterization tests were conducted on the latest sensor design at the University of Wyoming. It was found that the light-adaptation circuitry being utilized caused artifacts in the output signals. These artifacts, while initially bothersome, have a distinct use as a means to detect motion direction across the sensor’s field-of-view. This paper provides background information on the sensor design being considered, discusses one of the tests conducted, and offers results that illustrate the signal artifacts. The potential uses of these artifacts, as well as what causes them, is discussed in detail.

Musca domestica based machine vision sensor: a continuing project - biomed 2010.

The Musca domestica (common housefly) sensor project is an ongoing endeavor at the University of Wyoming. The project seeks to develop a sensor based on the rudimentary function of a flys eye, including the inherent hyperacuity (high sensitivity to motion) present in the insects vision system. The primary author participated in a Wyoming Undergraduate Experimental Program to Stimulate Competitive Research (EPSCoR) summer research program in 2009. He assisted other members of the research team and conducted independent research. Specifically, he designed light filter circuits, laid out complex circuit boards, developed tests for a new sensor design, practiced new design techniques, and developed hardware to aid in future testing. His involvement in the project expedited numerous processes vital to the projects ongoing success. Notably, he aided in the design and tested a newer, simpler sensor that implemented an unprecedented light filtering technology. He then designed and executed a number of tests on that sensor. This paper describes research accomplishments contributed through his EPSCoR summer research experience.