Space-time joint anomaly detection method based on hyperspectral image and electronic device

A hyperspectral image and anomaly detection technology, which is applied in the space-time joint anomaly detection method and the field of electronic equipment, can solve the problems of target detection omission or misjudgment, low signal-to-noise ratio, and inconsistency, and achieve low false alarm rate and high detection The effect of probability

CN108765460AActive Publication Date: 2018-11-06BEIHANG UNIV +1
7 Cites 5 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2018-11-06

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The present invention discloses a space-time joint anomaly detection method based on a hyperspectral image and an electronic device, the method comprising: acquiring the space anomaly map of each frame of image in a hyperspectral image sequence; acquiring the time anomaly map of each frame of image in the hyperspectral image sequence; acquiring a trajectory prediction map of the target detection of a current frame according to the joint anomaly map of a previous frame of image, wherein the previous frame is a frame adjacent to the current frame; according to the space anomaly map, the time anomaly map, and the trajectory prediction map of each frame, acquiring the joint anomaly map of a target in a hyperspectral image cube. The method has a low false alarm rate and a high detection probability in aerial aircraft target detection.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to image recognition technology, in particular to a hyperspectral image-based space-time joint anomaly detection method and electronic equipment. Background technique

[0002] With the development of hyperspectral sensors, hyperspectral images have been applied to many classic image processing problems. Compared with single-band sensors, hyperspectral sensors can simultaneously collect spectral feature information and spatial feature information of images. This is also an important advantage of hyperspectral images in the field of image processing.

[0003] Weak and small targets refer to targets of interest that are small in size, weak in intensity, and low in signal-to-noise ratio in the image. Weak and small object detection has a wide range of applications in both military and civilian fields, and has attracted the interest of many researchers. Due to the complex background, interference of noise clutter and attenuation of ...

Examples

experiment example

[0101] In order to evaluate the performance of the weak and small moving object detection method, a test dataset is composed of hyperspectral image sequences under complex cloud background. Such as Figure 4 and Figure 5 as shown, Figure 4 Shown is the effect of one band (wavelength 0.68 μm) in the hyperspectral image data sequence, Figure 5 What is shown is the three-dimensional projection effect of the image in this band. Due to the small size of the faint target, it is difficult to distinguish the target of interest from the background; and the cloud background and noise clutter will lead to a low signal-to-noise ratio.

[0102] Using the method of the embodiment of the present invention, the key frames to be detected are sequentially selected from the original sequence according to the fixed interval l=5. Annotate moving objects in keyframes for quantitative evaluation. In this experiment, the PCA principal component dimension k and the size of the convolution kerne...