Method and system for assessing health of a wood specimen

A wood health assessment system for assessing health of a wood specimen of a utility pole, the wood health assessment system comprising:

• a processor;
• and
• a memory communicatively coupled to the processor, wherein the memory stores processor-executable instructions, which on execution, cause the processor to:
• align, by an ultrasonic data capture and recording module, an ultrasonic device, comprising a transmitter and an associated receiver, to be positioned along a plurality of alignments, wherein the plurality of alignments being defined between a plurality of points;
• receive, by an ultrasonic data capture and recording module, ultrasonic data from the ultrasonic device for each of the plurality of alignments of the transmitter and the associated receiver across a cross-section of one or more cross-sections along a length of the wood specimen of the utility pole, wherein for each of the plurality of points, the ultrasonic data associated with each adjacent point from the plurality of points is excluded, and wherein the ultrasonic data comprises a pulse velocity, a transit time, and a distance travelled by an ultrasonic pulse between the transmitter and the associated receiver;
• measure, by a feature extraction module, relative features of the wood specimen of the utility pole using the ultrasonic data received for each alignment of the transmitter and the associated receiver, wherein the relative features comprise a mean of pulse velocities, a standard deviation of pulse velocities, a lower limit of pulse velocities, an upper limit of pulse velocities, a normalized lower limit of pulse velocities, and a normalized upper limit of pulse velocities;
• train a machine learning model using the relative features and at least one of a Support Vector Machine (SVM) model, a Random Forest model, and a Logical Regression model to generate a trained machine learning model;
• identify, by a defect detection module, a condition of the cross-section of the wood specimen of the utility pole based on the relative features of the wood specimen of the utility pole using the trained machine learning model;
• upon identifying the condition of the cross-section of the wood specimen of the utility pole to be a defective condition,
• determine, by a defect position determination module, a position of a defect in the cross-section of the wood specimen of the utility pole using the relative features of the wood specimen of the utility pole;
• and
• determine, by a defect severity determination module, a severity of the defect using the trained machine learning model and the relative features of the wood specimen of the utility pole, thereby, assessing the health of the wood specimen of the utility pole;
• and
• generate, by a wood health record generation module of the wood health assessment system, a health record for the wood specimen of the utility pole, wherein the health record comprises at least one of the condition of the cross-section of the wood specimen of the utility pole, the position of the defect in the cross-section of the wood specimen of the utility pole, the severity of the defect, and a visual representation of the position of the defect in the cross-section of the wood specimen of the utility pole.

A non-transitory computer readable medium including instructions stored thereon that when processed by at least one processor cause a wood health assessment system to perform operations comprising:

• aligning an ultrasonic device, comprising a transmitter and an associated receiver, to be positioned along a plurality of alignments, wherein the plurality of alignments being defined between a plurality of points;
• receiving ultrasonic data from the ultrasonic device for each of the plurality of alignments of the transmitter and the associated receiver across a cross-section of one or more cross-sections along a length of a wood specimen of a utility pole, wherein for each of the plurality of points, the ultrasonic data associated with each adjacent point from the plurality of points is excluded, and wherein the ultrasonic data comprises a pulse velocity, a transit time, and a distance travelled by an ultrasonic pulse between the transmitter and the associated receiver;
• measuring relative features of the wood specimen of the utility pole using the ultrasonic data received for each alignment of the transmitter and the associated receiver, wherein the relative features comprise a mean of pulse velocities, a standard deviation of pulse velocities, a lower limit of pulse velocities, an upper limit of pulse velocities, a normalized lower limit of pulse velocities, and a normalized upper limit of pulse velocities;
• training, by the wood health assessment system, a machine learning model using the relative features and at least one of a Support Vector Machine (SVM) model, a Random Forest model, and a Logical Regression model to generate a trained machine learning model;
• identifying a condition of the cross-section of the wood specimen of the utility pole based on the relative features of the wood specimen of the utility pole using the trained machine learning model;
• upon identifying the condition of the cross-section of the wood specimen of the utility pole to be a defective condition,
• determining a position of a defect in the cross-section of the wood specimen of the utility pole using the relative features of the wood specimen of the utility pole;
• and
• determining a severity of the defect using the trained machine learning model and the relative features of the wood specimen of the utility pole, thereby, assessing the health of the wood specimen of the utility pole;
• and
• generating, by a wood health record generation module, a health record for the wood specimen of the utility pole, wherein the health record comprises at least one of the condition of the cross-section of the wood specimen of the utility pole, the position of the defect in the cross-section of the wood specimen of the utility pole, the severity of the defect, and a visual representation of the position of the defect in the cross-section of the wood specimen of the utility pole.