Method and system for detection of a desired object occluded by packaging

Patent metadata

Field	Value
Publication number	US-12164017-B2
Application number	US-202117453103-A
Country	US
Kind code	B2
Filing date	Nov 1, 2021
Priority date	May 28, 2021
Publication date	Dec 10, 2024
Grant date	Dec 10, 2024

Abstract

An e-commerce business model has witnessed several cases where packages with faulty goods, returned by buyers, without procured object, rather replaced by different device. This disclosure relates a method to detect whether object under test a desired object. A plurality of back-scattered signals is received from the object under test occluded by packaging with continuous motion on conveyer based on first antenna-radar combination. The plurality of back-scattered signals is processed by applying four-tap difference filter to obtain motion-filtered data matrix. A low pass filter is applied on the motion-filtered data matrix to obtain enveloped motion-filtered data matrix. A sliding constant false alarm rate is applied on the enveloped motion-filtered data matrix to determine detection threshold value. A check is performed to detect whether the object under test is the desired object based on whether intensity of the plurality back-scattered signals exceeds the detection threshold value.

First claim

What is claimed is:

1. A processor implemented method for detecting a desired object from an object under test occluded by packaging, comprising:
scanning, via one or more hardware processors, the object under test packed inside a package positioned on a conveyor using ultra-wide band signals of an antenna-radar unit mounted on a scanning chamber;
receiving, via the one or more hardware processors, a plurality of back-scattered signals from the object under test by the antenna-radar unit-; and
processing, via the one or more hardware processors, the plurality of back-scattered signals using a first approach to detect whether the object under test is the desired object, wherein the first approach is applied when the object under test is having continuous motion across the scanning chamber, wherein processing via the first approach comprises:
(a) processing, the plurality of back-scattered signals by applying a four-tap difference filter to obtain a motion-filtered data matrix;
(b) applying, a low pass filter, on the motion-filtered data matrix to obtain an enveloped motion-filtered data matrix;
(c) applying, a sliding constant false alarm rate (CFAR) on the enveloped motion-filtered data matrix to determine a detection threshold value, wherein the detection threshold value is determined based on a scale factor (α), and wherein the scale factor (α) is a constant and selected based on a desired probability of a false alarm; and
(d) detecting, whether the object under test is the desired object based on whether an intensity of the plurality back-scattered signals exceeds the detection threshold value, wherein the false alarm occurs when the intensity of the plurality back-scattered signals exceeds the detection threshold value at a time and a range point where no desired object is located, wherein the detection threshold value is determined based on a plurality of average values of at least one set for at least one reference cell from set of reference cells, and a scale factor (α), and wherein the at least one cell corresponds to at least one guard cell which are filtered to remove a part of the actual peak value and wherein one or more guard cells, each said at least one guard cell representing a range bin that are immediately adjacent to every cell under test (CUT) are excluded from calculation to avoid corrupting estimate of a surrounding noise floor level with a reflected power (I) from an object under test, and a particular cell is chosen from the CUT, and a minimum (LO) of averages of reference cells present on both sides of the CUT, is multiplied by a scale factor (α) to find the detection threshold value (Th).

2. The processor implemented method as claimed in claim 1 , processing, via the one or more hardware processors, the plurality of back-scattered signals using a second approach to detect whether the object under test is the desired object, wherein the second approach is applied when the object under test having a plurality of stops across the scanning chamber, wherein processing via the second approach comprises:
converting, the plurality of back-scattered signals into a plurality of frequency domain signals by a Fast Fourier transform (FFT), wherein the plurality of back-scattered signals are acquired as a range-time matrix, wherein each of the plurality of frequency domain signals comprises (a) an amplitude component, and (b) a phase component;
extracting, via the one or more hardware processors, a plurality of features from the plurality of frequency domain signals at a first frequency band (f 0 ) and a second frequency band (2f 0 ), wherein the plurality of features captures signal patterns in the plurality of frequency domain signals; and
processing, by a binary classifier, the extracted plurality of features to detect whether the object under test is the desired object, wherein the binary classifier corresponds to a K nearest neighbour.

3. The processor implemented method as claimed in claim 1 , wherein the antenna-radar unit comprises a first antenna-radar combination having three radar units, wherein the three radar units comprises three antennas connected to three radar units respectively, wherein the first antenna-radar combination scans the object under test from an entry position to an exit position across the scanning chamber, and wherein the three radar units are time synchronized with a maximum uncertainty of +/−500 milli seconds.

4. The processor implemented method as claimed in claim 2 , wherein the antenna-radar unit comprises a second antenna-radar combination having three antennas connected to a radar unit, and wherein the second antenna-radar combination scans the object under test in a sequential manner at the plurality of stops across the scanning chamber.

5. A system for detection of a desired object from an object under test occluded by packaging, comprising:
a memory storing instructions;
one or more communication interfaces; and
one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to:
scan, the object under test packed inside a package positioned on a conveyer using ultra-wide band signals of an antenna-radar unit mounted on a scanning chamber;
receive, a plurality of back-scattered signals from the object under test by the antenna-radar unit; and
process, the plurality of back-scattered signals using a first approach to detect whether the object under test is the desired object, wherein the first approach is applied when the object under test is having continuous motion across the scanning chamber, wherein process via the first approach comprises:
(a) process, the plurality of back-scattered signals by applying a four-tap difference filter to obtain a motion-filtered data matrix;
(b) apply, a low pass filter, on the motion-filtered data matrix to obtain an enveloped motion-filtered data matrix;
(c) apply, a sliding constant false alarm rate (CFAR) on the enveloped motion-filtered data matrix to determine a detection threshold value, wherein the detection threshold value is determined based on a scale factor (a), and wherein the scale factor (a) is a constant and selected based on a desired probability of a false alarm; and
(d) detect, whether the object under test is the desired object based on whether an intensity of the plurality back-scattered signals exceeds the detection threshold value, wherein the false alarm occurs when the intensity of the plurality back-scattered signals exceeds the detection threshold value at a time and a range point where no desired object is located, wherein the detection threshold value is determined based on a plurality of average values of at least one set for at least one reference cell from set of reference cells, and a scale factor (α), and wherein the at least one cell corresponds to at least one guard cell which are filtered to remove a part of the actual peak value and wherein one or more guard cells, each said at least one guard cell representing a range bin that are immediately adjacent to every cell under test (CUT) are excluded from calculation to avoid corrupting estimate of a surrounding noise floor level with a reflected power (I) from an object under test, and a particular cell is chosen from the CUT, and a minimum (LO) of averages of reference cells present on both sides of the CUT, is multiplied by a scale factor (α) to find the detection threshold value (Th).

6. The system as claimed in claim 5 , wherein the one or more hardware processors are further configured by the instructions to process, the plurality of back-scattered signals using a second approach to detect whether the object under test is the desired object, wherein the second approach is applied when the object under

Assignees

• Tata Consultancy Services Ltd

Inventors

• Ray Arindam
• Das Taniya
• Chowdhury Arijit
• Chakravarty Soumya
• Rani Smriti
• Khasnobish Anwesha
• Chakravarty Tapas
• Pal Arpan

Classifications

• G01S13/87

  Combinations of radar systems, e.g. primary radar and secondary radar · CPC title

• G01S13/887

  for detection of concealed objects, e.g. contraband or weapons · CPC title

• G01S5/0226

  Transmitters · CPC title

• G01S13/89

  for mapping or imaging · CPC title

• G01S13/04 Primary

  Systems determining presence of a target (based on relative movement of target G01S13/56) · CPC title