Assessing Landmine Detection Using Geophysical Equipment in Sand and Soil Beds Similar to Donetsk Region of Eastern Ukraine
Landmines have undergone many technological advances post WWII. However, detection methods such as prodding or metal detectors have not adapted to these changes over the past half century, and are thereby outdated. Current research is becoming focused on using geophysical methods in landmine detection. Robotically-controlled geophysical equipment is adaptable, providing a reliable way to demine post-war zones and reduce human casualties. This study has a specific focus on evaluating potential methods demining the Donetsk region in Eastern Ukraine, using test beds in Lancaster PA to assess the effectiveness of combined geophysical methods in different soil types. These methods include holographic radar, impulse radar, metal detection and infrared imaging. These were used to locate buried mines and clutter in ten sand beds and ten soil beds. In the spring of 2016, images generated by these methods were presented to 37 minimally-trained subjects who were asked to identify the objects and label them as either mines or clutter. The results were recorded and used to generate a receiver operator characteristic (ROC) curve relating probability of detection (PD) to probability of false alarms (PFA). In both sand (53%) and soil (60%) beds, more than half of the mines were detected. Landmines in soil beds were detected more frequently than in sand beds. Large anti-tank mines were more easily identified than the smaller anti-personnel mines. Subjects rated the infrared images as most uninformative while the impulse images were rated the most helpful in discriminating landmines. Overall, these results provide the necessary preliminary baseline information for further field testing in Eastern Ukraine in the upcoming months. The goal of this experiment was not to arrive at perfect survey results, but rather to confirm that the equipment is suitable for use in field conditions similar to Eastern Ukraine, and by relatively non-expert personnel.
Franklin and Marshall College Archives, Undergraduate Honors Thesis 2016
- F&M Theses Collection