Unmanned aircraft system (UAS) detection and assessment via temporal intensity aliasing
A method and system for temporal frequency analysis for identification of unmanned aircraft systems. The method includes obtaining a sequence of video image frames and providing a pixel from an output frame of the video; generating a fluctuating pixel value vector; examining the fluctuating pixel value vector over a period of time; obtaining the frequency information present in the pixel fluctuations; summing the frequency coefficients for the vectorized pixel values from the fluctuating pixel value vector; obtaining an image representing a two dimensional space based on the summed center frequency coefficients; generating a series of still frames equal to a summation of the center frequency coefficients for pixel variations; and combining the temporal information into spatial locations in a matrix to provide a single image containing the spatial and temporal information present in the sequence of video image frame.
1. A system for temporal frequency analysis for identification of an unmanned aircraft system (UAS) comprising a data collection unit, a graphical processing unit and an output device in data communication;
the data collection unit comprising a video camera configured to acquire a sequence of video data associated with an area of interest; wherein
the GPU comprising a temporal frequency analysis module, a machine learning module, and a classification module in data communication; and
the graphical processing unit being configured to collect data transmitted to the graphical processing unit (GPU) via the data collection unit; process the video data in a temporal sequence of frames via the graphical processing unit; perform analysis of temporal frequency of the respective sequence of frames via the temporal frequency analysis module; apply at least one machine learning algorithm to the analyzed frames via the machine learning module; classify the temporal frequency analysis via the classification module; and transmit the processed video data to the output device; and
wherein the temporal frequency analysis module is configured to process the temporal data using a Fourier transform algorithm comprising:
B
_
=
∑
i
=
1
n
B
(
i
)
n
Δ
B
_
(
t
)
=
∑
n
,
m
=
1
k
B
(
n
,
m
,
t
)
-
B
_
k
Δ
T
_
(
t
)
=
∑
n
,
m
=
1
k
T
(
n
,
m
,
t
)
-
B
_
k
wherein B′ is the average background image, ΔB′ is the average of the changes in each background frame, ΔT′ is the averages of the changes in each object data frame, B is the background frame data, and T is the target frame data.
2. The system of claim 1 , wherein the at least one machine learning algorithm comprises a convoluted neural network (CNN), and applying the CNN to the data output of the temporal frequency analysis module.
3. The system of claim 2 , wherein the classification module performs a classification based on an output of the temporal frequency analysis module and the original video data.
4. The system of claim 3 , wherein the output device comprises a video data management system in data communication with an operator station; the video data management system configured for viewing and analysis of output data generated by the GPU.
5. The system of claim 1 , wherein the temporal frequency analysis module is further configured to perform the fast Fourier transform algorithm based on an object data gradient frequency spectrum and a background data gradient defined by an algorithm:
=
(
∇
Δ
)
=
(
∇
Δ
)
wherein is the frequency spectrum of the gradient of averages of the object data and is the frequency spectrum of the gradient of averages of the background data.