We present a TensorFlow implementation of the RX-algorithm for anomaly detection in multi-spectral and hyperspectral imagery. In this paper, we perform a runtime performance comparison of the algorithm, implemented us...
详细信息
ISBN:
(纸本)9781510666955;9781510666962
We present a TensorFlow implementation of the RX-algorithm for anomaly detection in multi-spectral and hyperspectral imagery. In this paper, we perform a runtime performance comparison of the algorithm, implemented using: NumPy, SciPy and TensorFlow libraries on a CPU, a GPU (where applicable) as well as on edge hardware (Jetson TX2). The RX detection algorithm makes use of either local or global background statistics, such as the mean and covariance, to find anomalous pixels. In the approach examined here, the statistics are estimated using local background samples from the area neighbouring the pixel under test. Such algorithms are typically implemented in Python using the NumPy library for numerical operations;however, a preliminary literature review found no formal investigations have been made into the suitability of alternative frameworks to optimise the performance on edge hardware. Our TensorFlow (and SciPy) implementations involve the use of a convolutional operations to calculate the required statistics. The use of these libraries significantly reduces the algorithm's run time. We evaluate the implementation using a range of hardware, in order to get a diverse set of results and to highlight the differences in run times on each. We also show a comparative set of implementations of a matched filter algorithm for target detection. This algorithm uses a very similar approach to the RX algorithm, but is provided with a template target spectrum to detect within the image. Notable improvements (similar to 98% reduction in run time) in performance can be seen through the use of a TensorFlow implementation on GPU. Results are demonstrated by trialing on multispectral imagery for ship detection.
In order to enhance the positioning capability of terrestrial networks, a novel communication and navigation fusion signal is proposed. The novel signal multiplexes the communication and navigation signal in the same ...
详细信息
In order to enhance the positioning capability of terrestrial networks, a novel communication and navigation fusion signal is proposed. The novel signal multiplexes the communication and navigation signal in the same frequency band, and the navigation system is superimposed on the original communication system. However, the application of pseudorandom noise (PRN) sequences in the navigation system is limited by the communication clock period. Taking the application of PRN sequences limited by the clock period as objects, the present study analyzes truncated PRN (TPRN) sequences. PRN sequences with a TPRN sequence as the navigation signal can overcome the communication system clock period limitation. Then, a matched filter algorithm with double detection (MFADD) is proposed to acquire the novel signal. The matchedfilter method is applied to the proposed algorithm to determine the start code phase of TPRN. Monte Carlo simulations and real data tests demonstrate the effectiveness of the proposed algorithm for the designed signal.
暂无评论