Search

Ground Penetrating Radar (GPR) is a non-destructive test used as imaging tool for exploration of shallow subsurface such locating the buried infrastructures. Due to the existence of various subsurface material and environmental noise, such as bricks and tree branch, it is a challenging task to interpret the GPR data into a meaningful information. Thus, this project proposes the hyperbolic recognition of buried pipes in GPR images in the presence of scattering objects. In this framework, the GPR images were firstly subjected to image pre-processing. Then, the GPR images were decomposed using Discrete Wavelet Transform (DWT) to analyze the texture analysis of hyperbola signature. Then, the approximation subband of DWT were extracted and used as features to recognize the hyperbolic of buried pipes and scattering objects presence in GPR images. A series of experiment has been conducted on GPR data collection at Agency Nuclear Malaysia. Based on the results obtained, the average recognition rate of extracted approximation subband of DWT features using k-NN classifier is 99.75%, thus shows a promising results in recognizing the buried pipes in the presence of scattering objects. [ABSTRACT FROM AUTHOR]

Ground Penetrating Radar (GPR) is a geophysical locating method that uses radio waves to capture images below the surface of the ground in a minimally invasive way. It also requires two main essential equipment which is a transmitter and a receiving antenna. To address the problem, this project proposed the hyperbolic recognition of different utility pipes of GPR images at different level of penetration depth. In this framework, the raw data of GPR images were firstly to be pre-processed. The grayscale images were cropped, resized, and enhanced to increase the contrast of the features of the image. Then, the pre-processed GPR images were extracted using the Histogram of Oriented Gradient (HOG) method with three different windows. The extracted HOG features were then used as input to the k-Nearest Neighbor classifier. A series of experiments has been conducted using 10-fold cross-validation technique for training and testing the GPR data. Based on the result obtained, it shows that at depth 20 cm the average accuracy is about 99.87%, whereas at depth 40 cm the average accuracy achieved 100%. Thus, the result shows that the extracted HOG features exhibit the significant information of hyperbolic signature of different pipe size with different depth of buried object. Therefore the results seem promising in recognizing the hyperbolic of utilities. [ABSTRACT FROM AUTHOR]

Ground Penetrating Radar (GPR) is a geophysical method using high resolution electromagnetic used to acquire the information of underground. The electromagnetic (EM) waves produces from the antenna consisting of transmitter and receiver. The waves from the transmitter penetrates into the ground and reflect backs to the surface that receive by the antenna receiver. The antenna can lie within the range of 10MHz to 1000MHz to determine the shallow or deep penetration. Higher value of antenna will result in shallow penetration and otherwise for lower antenna. The process of recognition of buried objects is challenging task especially in the construction area to ensure safety and the quality of civil building. The GPR will display the mapping image on its control unit screen. If there are objects underground have detected, the image will display the hyperbola shape to indicate the target of the object. A vast number of data makes it difficult to classify each and every one of it either the image data is in which classes or categories. If there are many hyperbola present in image also makes it difficult to locate the accurate position. Due to this, deep learning technique by means of ResNet50 has been used in this research for hyperbola recognition in GPR image. A series of experiments has been conducted on the GPR dataset collected at Agency Nuclear Malaysia. Based on the results obtained, the deep learning model successfully learn the image feature. The accuracy of the model classified for this GPR data using ResNet50 gives 90% accuracy. Therefore, the proposed method for image recognition shows the promising results with all the GPR images are correctly recognize. Further, region of interest of hyperbola signature has been represented by a rectangular box indicates the hyperbola location [ABSTRACT FROM AUTHOR]

Ground penetrating radar has been acknowledged as an effective and efficient technique for non-destructive investigation for near-subsurface exploration that is based on the reflection receiver-transmitter of the antenna when hitting buried objects. An accurate interpretation of GPR data is greatly important in locating and mapping underground objects. Although GPR research has achieved remarkable success, the interpretation of GPR raw data highly depends on the reliance of user experts. Further, unexperienced GPR users are subject to error since the hyperbola signatures may resemble each other. Therefore, this work focuses on the development of a 3D reconstruction of the hyperbola signature of underground objects using GPR images for mapping applications. In this study, 3D reconstruction has been developed based on the Synthetic Aperture Focusing Technique, also known as SAFT. At the first stage, the raw input of GPR images was subjected to zero-time correction and background elimination. Next is the projection of each hyperbola signature by means of B-Scan images to create a 3D image. Then, the resultant 3D images were stacked together, and further 3D interpolation techniques were employed on the images. The experimental studies have been done on GPR data using a metal sphere as a sample. The findings of the study highlight that the SAFT method was able to reconstruct the 3D model of the hyperbola signature and exhibit the ability to provide clues about the location of the underground object through the representation of the voxel point of the images. Based on these results, the SAFT technique provides good insight into the 3D reconstruction of hyperbola signatures using GPR images in mapping applications. [ABSTRACT FROM AUTHOR]