Building detection by local region features in SAR images

Abstract

The buildings are very complex for detection on SAR images, where the basic features of those are shadows. There are many different representations for SAR shadow. As result it is no possible to use convolutional neural network for building detection directly. In this article we give property analysis of SAR shadows of different type buildings. After that, each region (ROI) prepared for training of building detection is corrected with its own SAR shadow properties. Reconstructions of ROI will be put in a modified YOLO network for building detection with better quality result.

Fast matrix multiplication for binary and ternary CNNs on ARM CPU

“…Although the number of instructions (COM, LD, MOV, and INS in Table II) can give us general ideas on which algorithm should have better computational efficiency, in practice the efficiency also depends on cache misses (which in turn depends on the order of loads and stores in memory), the ability of CPU to use the instruction pipelining (that depends on the order in which instructions are fetched). Furthermore, the overall efficiency of the matrix multiplication is affected also by reordering operations that prepare matrix blocks for microkernel and by post-processing in algorithms U8 and U4, which is shown in (3). That is why we experimentally measure the efficiency of all the algorithms under consideration.…”

“…Convolutional neural networks (CNNs) are the primary tool for solving various computer vision problems: pattern recognition [1], detection of different objects [2], [3], semantic segmentation [4] and many others. Although new transformerbased [5], [6] or deep MLP-based [7], [8] neural networks sometimes outperform CNNs on challenging datasets, they are usually harder to train, have more parameters and require more computational resources for inference [8], [9].…”