Fig 8 - available via license: CC BY
Content may be subject to copyright.
Samples of borehole-wall images. (A) border images; (B) fracture images; (C) intact rock mass images. https://doi.org/10.1371/journal.pone.0199749.g008
Source publication
Analyzing geological drilling hole images acquired by Axial View Panoramic Borehole Televiewer (APBT) is a key step to explore the geological structure in a geological exploration. Conventionally, the borehole images are examined by technicians, which is inefficient and subjective. In this paper, three dominant types of borehole-wall images on coal...
Similar publications
In the present paper, computer-aided system for the early detection of pulmonary nodules in Computed Tomography (CT) scan images is developed where pulmonary nodules are one of the critical notifications to identify lung cancer. The proposed system consists of four main stages. First, the raw CT chest images were preprocessed to enhance the image c...
![Fig. 1: The basic LBP operator on a 3 × 3 window [19]](profile/Dhruv-Sharma-9/publication/326901713/figure/fig1/AS:696659864453120@1543108063447/The-basic-LBP-operator-on-a-3-3-window-19_Q320.jpg)
Encoded Local Projections (ELP) is a recently introduced dense sampling image descriptor which uses projections in small neighbourhoods to construct a histogram/descriptor for the entire image. ELP has shown to be as accurate as other state of the art features in searching medical images while being time and resource efficient. This paper attempts...
Presented herein is a vision-based method for the detection of anomalies on roadway pavements, utilizing low-cost video acquisition and image processing of road surface frames collected by a smartphone (or camera) located on a vehicle moving in a real-life urban network, along with entropy-based texture segmentation filters, and support vector mach...
In the past decade, face recognition is considered as an important biometric type due to its wide applications in practice in terms of authentication. The verification process of a human face is not trivial task especially different face poses are captured to be matched. In this paper, an efficient algorithm for face recognition is proposed. In the...
![Fig.4: Reference figures [3] (a) Original Image 8x8 size (b) second...](publication/326253827/figure/fig2/AS:710709151297542@1546457674826/Reference-figures-3-a-Original-Image-8x8-size-b-second-level-c-final-level-8x8_Q320.jpg)
Synthetic Aperture Radar (SAR) is not only having the characteristic of obtaining images during all-day, all-weather, but also provides object information which is distinctive from visible and infrared sensors. but, SAR images have more speckles noise and fewer bands. This paper propose a method for denoising, feature extraction and classification...
Citations
... Each classifier consists of an extended ResNet-50 two-stage CNN. We use two-stage CNNs herein as they have been shown to achieve higher detection and classification accuracy compared to single-stage CNNs [63], as well as being able to incorporate robust feature extraction when subtle variations exist [64]. ...
... compared to single-stage CNNs [63], as well as being able to incorporate robust feature extraction when subtle variations exist [64]. ...
Wearing masks in indoor and outdoor public places has been mandatory in a number of countries during the COVID-19 pandemic. Correctly wearing a face mask can reduce the transmission of the virus through respiratory droplets. In this paper, a novel two-step deep learning (DL) method based on our extended ResNet-50 is presented. It can detect and classify whether face masks are missing, are worn correctly or incorrectly, or the face is covered by other means (e.g., a hand or hair). Our DL method utilizes transfer learning with pretrained ResNet-50 weights to reduce training time and increase detection accuracy. Training and validation are achieved using the MaskedFace-Net, MAsked FAces (MAFA), and CelebA datasets. The trained model has been incorporated onto a socially assistive robot for robust and autonomous detection by a robot using lower-resolution images from the onboard camera. The results show a classification accuracy of 84.13% for the classification of no mask, correctly masked, and incorrectly masked faces in various real-world poses and occlusion scenarios using the robot.
... The classification technique is used for feedback processing in the form of OEQ because the feedback will be grouped on predetermined survey aspects. In particular, this research uses two classification algorithms, namely Support Vector Machine (SVM) [8][9] [10] and Multi-Layer Perception (MLP) [11] [12][13] [14]. ...
Borehole imaging is one of the main methods for fracture observation in rock formations, but there is still a lack of identification systems for deriving fracture parameters from borehole images. Additionally, previous research was limited to sinusoidal fractures. In this study, an intelligent system is developed for multiple types of qualitative fracture identification in a quick and robust manner using convolution neural networks and post-processing. Initially, borehole imaging was carried out in six coal mines to develop a borehole image database that includes a wide variety of rock types and fractures. Afterward, the acquired borehole images of fractures were categorized into three types, such as sinusoidal fractures, vertical fractures, and broken areas. In this connection, dual convolutional neural networks were developed to recognize fracture-type pixels. Meanwhile, the calculation models associated with corresponding fracture parameters were established to quantitatively describe the fracture characteristics. Eventually, the post-processing programs were developed to identify the geometric parameters for each fracture type. The results indicate that the developed dual convolutional neural networks have the capability to detect fracture pixels with a recognition accuracy of 77.1% for each sinusoidal and vertical fracture pixels, and a recognition accuracy of 72.9% for broken area pixels. The post-processing program was developed based on the Hough transform and a thinning algorithm, that enables it to reach an average recognition rate of 88.71% with an average false alarm rate of 16.67% for all three types of fractures.
Borehole imaging, one of the main methods of rock formation detection, is widely used in mining, tunnels, petroleum, and other fields of engineering. However, the method of manually identifying the characteristic of rock formation based on borehole imaging is greatly affected by subjective factors. For example, it should be noted that when the rock layer color is similar, the position of the rock interface may be different. In this paper, a method of quantitatively describing the characteristics of the rock formation is proposed, and the automatic recognition of the rock interface is realized. Firstly, the calculation method of rock dip angle and position based on borehole image with 360° wall surface development is proposed. Secondly, the three-dimensional distribution characteristics of the gray values of the rock layer are used to quantitatively describe the characteristics of the rock layer. Finally, an automatic recognition method of rock formation interface based on change-point detection algorithm is proposed. In addition , the recognition effect of the rock formation interface is analyzed by experiments and applied in the field. Research shows that, compared with method two, method one can determine the rock inclination. But its anti-interference ability is poor. Method two can better determine the position of the rock interface. The method is of great significance to the intelligent development of the borehole imaging system and promotes safe, efficient, and sustainable mining in coal mines.
