US-12618803-B2 - System and method for automated acquisition and analysis of electromagnetic testing data

Abstract

A method for identifying indications in an object via non-destructive testing using inspection equipment comprising a probe is described. The method includes: recording test data corresponding to a signal measurement acquired by the probe; processing the test data using a first analysis machine learning algorithm trained to output a list of detected landmarks; processing the list of detected landmarks to identify regions in the object based on the landmarks; processing the test data and the identified regions using a second analysis machine learning algorithm to output a list of detected indications; processing the list of indications to automatically classify each indication according to one of a plurality of predefined indication types; and outputting the classified indications in a report specifying positions of the classified indications in the object. A corresponding system and non-transitory computer-readable medium are also described.

Inventors

• Philippe MACKAY
• Vincent GAUDREAULT
• Florian Hardy
• Marco Michele Sisto

Assignees

• EDDYFI CANADA INC.

Dates

Publication Date

20260505

Application Date

20221027

Claims (20)

1. 1 . A method for identifying indications in an object via non-destructive testing using inspection equipment comprising a probe, the method comprising: recording test data from the probe while the probe is operated to scan the object, the test data comprising a plurality of data points, each data point corresponding to a signal measurement acquired by the probe; processing the test data using a first analysis machine learning algorithm trained to identify landmarks in the test data and output grouped test data identifying ranges in the test data corresponding to the identified landmarks; modelling the grouped test data as a chain of identified landmarks, and identifying patterns within the chain of identified landmarks corresponding to an expected sequence of landmarks; extracting an isolated landmark sequence comprising landmarks that match the patterns; processing the isolated landmark sequence to identify regions in the object based on the landmarks; processing the test data and the identified regions using a second analysis machine learning algorithm trained to detect indications in the test data and output a list of detected indications; processing the list of detected indications to automatically classify each of the indications according to one of a plurality of predefined indication types; and outputting each of the classified indications in a report specifying positions of each of the classified indications in the object.
2. 2 . The method according to claim 1 , comprising, prior to recording the test data: recording calibration data from the probe while the probe is operated to scan a reference object; processing the calibration data using a calibration machine learning algorithm trained to detect and identify reference signatures in the calibration data; providing the identified reference signatures to a calibration algorithm to extract calibration parameters; and calibrating the inspection equipment by applying the calibration parameters.
3. 3 . The method according to claim 2 , wherein the calibration machine learning algorithm is trained on historical calibration data comprising reference signatures labelled according to a plurality of possible indication types, and processing the calibration data comprises classifying reference signatures in the calibration data according to one of the plurality of possible indication types.
4. 4 . The method according to claim 3 , wherein the calibration machine learning algorithm is configured to calculate a confidence score corresponding to an estimated confidence level of reference signature classifications.
5. 5 . The method according to claim 4 , comprising outputting an error code when the confidence score is below a predetermined threshold.
6. 6 . The method according to claim 1 , comprising extracting normalization coefficients from the test data, and normalizing the test data by applying the extracted normalization coefficients prior to processing the test data using the first analysis machine learning algorithm.
7. 7 . The method according to claim 6 , wherein extracting normalization coefficients comprises processing the test data using a normalization machine learning algorithm trained to predict normalization coefficients, the normalization machine learning algorithm being trained using a regression algorithm on historical test data comprising corresponding normalization coefficients.
8. 8 . The method according to claim 7 , wherein the normalization machine learning algorithm is configured to calculate a confidence score corresponding to an estimated confidence level of normalization coefficient predictions.
9. 9 . The method according to claim 1 , wherein the second analysis machine learning algorithm comprises an object detection machine learning algorithm trained to recognize indications in the test data and directly output the list of detected indications including, for each indication, a start position and a stop position of data points in the test data that correspond to the indication.
10. 10 . The method according to claim 1 , wherein the second analysis machine learning algorithm is trained to segment indications in regions of the object identified using the identified landmarks identified by the first analysis machine learning algorithm.
11. 11 . The method according to claim 1 , wherein the first analysis machine learning algorithm is configured to output classified data points in which each of the plurality of data points is classified according to at least one of a plurality of predetermined landmark types, the method further comprising: processing the classified data points to generate a sequence of identified landmarks; and identifying regions in the object based on the sequence of identified landmarks.
12. 12 . The method according to claim 11 , wherein processing the classified data points comprises: detecting connected regions in the test data corresponding to sequences of the plurality of data points likely corresponding to a same type of the plurality of predetermined landmark types; generating the sequence of identified landmarks by concatenating each of the connected regions; comparing the sequence of identified landmarks to a model defining the expected sequence of landmarks in the object, and identifying a subset of identified landmarks in the sequence of identified landmarks that fits best with the model; and outputting the landmark sequence corresponding to the subset of identified landmarks.
13. 13 . The method according to claim 11 , comprising automatically validating the sequence of identified landmarks using one or more predetermined criteria, and requesting manual validation of the sequence of identified landmarks if the automatic validation fails.
14. 14 . The method according to claim 1 , wherein the first analysis machine learning algorithm configured to calculate a confidence score corresponding to an estimated confidence level of the identified landmarks.
15. 15 . The method according to claim 1 , comprising: processing the test data and the identified regions of the object using the second analysis machine learning algorithm, the second analysis machine learning algorithm being configured to output segmented data points in which each of the plurality of data points is classified as corresponding to an indication or not corresponding to an indication; and processing the segmented data points to generate a list of detected indications.
16. 16 . The method according to claim 15 , wherein processing the segmented data points comprises identifying connected regions in the segmented data corresponding to sequences of data points likely corresponding to a same indication.
17. 17 . The method according to claim 15 , wherein the second analysis machine learning algorithm is configured to calculate a confidence score corresponding to an estimated confidence level of detected indications.
18. 18 . The method according to claim 1 , wherein automatically classifying each of the indications comprises applying a predefined decision tree to either discard the indication or classify the indication according to one of a plurality of predefined indication types.
19. 19 . A system for identifying indications in an object via non-destructive testing, the system comprising: a probe; a recording device in operative communication with the probe, the recording device comprising a storage module configured to record and store test data from the probe while the probe is operated to scan the object, the test data comprising a plurality of data points, each data point corresponding to a signal measurement acquired by the probe; and an analysis device configured to access the test data stored by the recording device, the analysis device comprising a test data analysis module configured to: process the test data using a first analysis machine learning algorithm trained to identify landmarks in the test data and output grouped test data identifying ranges in the test data corresponding to the identified landmarks; model the grouped test data as a chain of identified landmarks, and identify patterns within the chain of identified landmarks corresponding to an expected sequence of landmarks; extract an isolated landmark sequence comprising landmarks that match the patterns; process the isolated landmark sequence to identify regions in the object based on the landmarks; process the test data and the identified regions using a second analysis machine learning algorithm trained to detect indications in the test data and output a list of detected indications; process the list of detected indications to automatically classify each of the indications according to one of a plurality of predefined indication types; and output each of the classified indications in a report specifying positions of each of the classified indications in the object.
20. 20 . A non-transitory computer-readable medium having instructions stored thereon to identify indications in an object via non-destructive testing using inspection equipment comprising a probe, the instructions, when executed by one or more processors, cause the one or more processors to: record test data from the probe while the probe is operated to scan the object, the test data comprising a plurality of data points, each data point corresponding to a signal measurement acquired by the probe; process the test data using a first analysis machine learning algorithm trained to identify landmarks in the test data and output grouped test data identifying ranges in the test data corresponding to the identified landmarks; model the grouped test data as a chain of identified landmarks, and identifying patterns within the chain of identified landmarks corresponding to an expected sequence of landmarks; extract an isolated landmark sequence comprising landmarks that match the patterns; process the isolated landmark sequence to identify regions in the object based on the landmarks; process the test data and the identified regions using a second analysis machine learning algorithm trained to detect indications in the test data and output a list of detected indications; process the list of detected indications to automatically classify each of the indications according to one of a plurality of predefined indication types; and output each of the classified indications in a report specifying positions of each of the classified indications in the object.

Description

CROSS REFERENCE TO RELATED APPLICATION This application claims the benefit of and priority to U.S. Provisional Patent Application No. 63/263,269, filed Oct. 29, 2021, and entitled SYSTEM AND METHOD FOR AUTOMATED ACQUISITION AND ANALYSIS OF ELECTROMAGNETIC TESTING DATA, the entirety of which is incorporated herein by reference. TECHNICAL FIELD The technical field generally relates to non-destructive testing, and more specifically to systems and methods for testing ferromagnetic and non-ferromagnetic tubes, which are often inspected by inserting probes in the tube. BACKGROUND Non-destructive testing (NDT) is the process of inspecting objects, without inducing permanent modification of the object, with the aim of identifying defects, imperfections, or other meaningful physical features of the object (referred to as “indications”). Different NDT techniques exist that employ one or more electromagnetic or acoustic sensors in array configurations, including but not limited to: Eddy Current Testing (ECT), Eddy Current Arrays (ECA), ECT rotating probes, Remote Field Testing (RFT), Remote Field Arrays (RFA), Internal Rotary Inspection Systems (IRIS), Partially Saturated Eddy Current Testing (PSECT), Fully Saturated Eddy Current Testing (FSECT), Magnetic Flux Leakage (MFL), Near Field Testing (NFT), Near Field Arrays (NFA), and Pulsed Eddy Current (PEC). Regardless the NDT technique, the inspection is performed by capturing electromagnetic/acoustic “raw” data with probes and instruments. The data is then normalized and analyzed by a human analyst to identify all interesting indications. Both the acquisition and the analysis processes can be long and error prone. There is therefore a need in the industry for solutions to assist probe operators in the acquisition of good quality data and assist analysts in the detection and classification of indications. SUMMARY According to an aspect, a method is provided for identifying landmarks in an object from signals captured by a probe during non-destructive testing of the object. The method includes: processing the signals using a machine learning algorithm to identify a plurality of predetermined landmark types, the algorithm being configured to output a sequence of connected regions in the test data corresponding to sequences of the plurality of data points likely corresponding to a same type of the plurality of predetermined landmark types; generating a sequence of identified landmarks by concatenating each of the connected regions; comparing the sequence of identified landmarks to a model defining an expected sequence of landmarks in the object, and identifying a subset of identified landmarks in the sequence of identified landmarks that fits best with the model; and outputting a landmark sequence corresponding to the subset of identified landmarks. According to an aspect, a method for identifying indications in an object via non-destructive testing is provided. The method includes: recording test data from a probe operated to scan the object, the test data comprising a plurality of data points, each data point corresponding to a signal measurement acquired by the probe; processing the test data using a first machine learning algorithm trained to identify a plurality of predetermined landmark types, the first algorithm being configured to output a sequence of identified landmarks and to identify regions in the object based on the landmarks; processing the test data and the identified object regions using a second machine learning algorithm trained to segment indications in the identified regions of the object from the test data, the second neural network being configured to output a list of detected indications; for each indication in the list of detected indications, applying a predefined decision tree to either discard the indication or classify the indication according to one of a plurality of predefined indication types; and outputting the classified indications in a report specifying positions of the classified indications relative to at least some of the identified landmarks. According to an aspect, a method is provided for identifying landmarks in an object from signals captured by a probe during non-destructive testing of the object. The method includes: processing the signals using a neural network trained to identify a plurality of predetermined landmark types, the algorithm being configured to output a sequence of connected regions in the test data corresponding to sequences of the plurality of data points likely corresponding to a same type of the plurality of predetermined landmark types; generating a sequence of identified landmarks by concatenating each of the connected regions; comparing the sequence of identified landmarks to a model defining an expected sequence of landmarks in the object, and identifying a subset of identified landmarks in the sequence of identified landmarks that fits best with the model; and outputting a landmark sequence corresponding