No abstract

A common countermeasure to detect threatening drones is the electro-optical infrared (EO/IR) system. However, its performance is drastically reduced in conditions of complex background, saturation and light reflection. 3D laser sensor LiDAR is used to overcome the problems of 2D sensors like EO/IR, but it is not enough to detect small drones at a very long distance because of low laser energy and resolution. To solve this problem, A 3D LADAR sensor is under development. In this work, we study the detection methodology adequate to the LADAR sensor which can detect small drones at up to 2 km. First, a data augmentation method is proposed to generate a virtual target considering the laser beam and scanning characteristics, and to augment it with the actual LADAR sensor data for various kinds of tests before full hardware system developed. Second, a detection algorithm is proposed to detect drones using voxel-based background subtraction and variable radially bounded nearest neighbor (V-RBNN) method. The results show that 0.2 m L2 distance and 60% expected average overlap (EAO) indexes are satisfied for the required specification to detect 0.3 m size of small drones.

Unmanned aerial vehicles and battleships are equipped with the infrared search and tracking (IRST) systems for its mission to search and detect targets even in low visibility environments. However, infrared sensors are easily affected by diverse types of conditions, therefore most of IRST systems need to apply advanced contrast enhancement (CE) methods to cope with the low dynamic range of sensor output and image saturation. The general histogram equalization for infrared images has unwanted side effects such as low contrast expansion and saturation. Also, the local area processing for saturation reduction has been studied to solve the problems regarding the saturation and non-uniformity. We propose the cross fusion based adaptive contrast enhancement with three counter non-uniformity methods. We evaluate the proposed method and compare it with conventional CE methods using the discrete entropy, PSNR, SSIM, RMSE, and computation time indexes. We present the experimental results for images from various products using several datasets such as infrared, multi-spectral satellite, surveillance, general gray and color images, as well as video sequences. The results are compared using the integrated image quality measurement index and they show that the proposed method maintains its performance on various degraded datasets.

The article presents a method for segmentation of ethnomusicological field recordings. Field recordings are integral documents of folk music performances captured in the field, and typically contain performances, intertwined with interviews and commentaries. As these are live recordings, captured in non-ideal conditions, they usually contain significant background noise. We present a segmentation method that segments field recordings into individual units labelled as speech, solo singing, choir singing, and instrumentals. Classification is based on convolutional deep networks, and is augmented with a probabilistic approach for segmentation. We describe the dataset gathered for the task and the tools developed for gathering the reference annotations. We outline a deep network architecture based on residual modules for labelling short audio segments and compare it to the more standard feature based approaches, where an improvement in classification accuracy of over 10% was obtained. We also present the SeFiRe segmentation tool that incorporates the presented segmentation method.

Fig. 1: The image shows a volume containing several intact SARS-CoV-2 virions acquired using cryo-electron tomography 3D imaging. From left to right: slice of the original data; direct volume rendering of the original data; foreground-background segmentation; color-coded four-class segmented data (background, spikes, membrane, lumen).