Speeded Up Robust Features Research Articles

Efficient Forgery Detection Approaches for Digital Color Images

This paper is concerned with a vital topic in image processing: color image forgery detection. The development of computing capabilities has led to a breakthrough in hacking and forgery attacks on signal, image, and data communicated over networks. Hence, there is an urgent need for developing efficient image forgery detection algorithms. Two main types of forgery are considered in this paper: splicing and copy-move. Splicing is performed by inserting a part of an image into another image. On the other hand, copy-move forgery is performed by copying a part of the image into another position in the same image. The proposed approach for splicing detection is based on the assumption that illumination between the original and tampered images is different. To detect the difference between the original and tampered images, the homomorphic transform separates the illumination component from the reflectance component. The illumination histogram derivative is used for detecting the difference in illumination, and hence forgery detection is accomplished. Prior to performing the forgery detection process, some pre-processing techniques, including histogram equalization, histogram matching, high-pass filtering, homomorphic enhancement, and single image super-resolution, are introduced to reinforce the details and changes between the original and embedded sections. The proposed approach for copy-move forgery detection is performed with the Speeded Up Robust Features (SURF) algorithm, which extracts feature points and feature vectors. Searching for the copied partition is accomplished through matching with Euclidian distance and hierarchical clustering. In addition, some pre-processing methods are used with the SURF algorithm, such as histogram equalization and single-mage super-resolution. Simulation results proved the feasibility and the robustness of the pre-processing step in homomorphic detection and SURF detection algorithms for splicing and copy-move forgery detection, respectively.

Fusion of Novelty Detectors Using Deep and Local Invariant Visual Features for Inspection Task

In this study, a novel framework using multiple novelty detection filters is developed to learn a model of the normality of a robot’s visual perception, which is called multichannel novelty detection. Subsequently, the acquired model was used to highlight dissimilar perceptions when the robot explored an environment. The main purpose of fusing multiple novelty filters is that each novelty filter performs well in detecting specific types of novelties; therefore, a new framework is proposed that demonstrates a new way to combine multiple different purposed novelty detection filters together in order to yield an overall more robust novelty status on the visual features. To develop a multichannel novelty detection system, expectation- and appearance-based novelty detection models were used in this study. To become experts in detecting different types of novelty using these models, different features from the input image were extracted as inputs for the models. The expectation-based novelty detection model uses the MobileNetV2 deep network to extract the deep features of the input image, which is subsequently used to learn a sequential and temporal model of normality to detect novelty. By contrast, the appearance-based novelty detection model uses speeded up robust features (SURF), which provide more region-focused features within the input image, to identify whether a specific region of the image is novel. The proposed multichannel novelty detection system is a completely online and real-time approach that is very important for mobile robotics applications. The proposed framework was tested in three novel environments, and it was reported that the proposed multichannel novelty detection system performs better than expectation-based and appearance-based novelty filters separately. Statistically, in the three novel environments, the Matthews correlation coefficients are reported to be 0.94, 0.97, and 0.93, and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</i> <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> scores are reported to be 0.95, 0.97, and 0.93, respectively, which proves that and can be concluded as almost perfect statistically.

Study of Keypoints Detectors and Descriptors Performance on X-Ray Images Compared to the Visible Light Spectrum Images

In this work, we study the performance of wide-used keypoints detection and description algorithms: Scale-Invariant Feature Transform (SIFT), Speeded Up Robust Features (SURF), Oriented FAST and Rotated BRIEF (ORB), Binary Robust Invariant Scalable Keypoints(BRISK), Accelerated KAZE(AKAZE), which were originally developed for images taken in visible light but widely applied in the fields where images are taken in a different spectrum. We compare the quality of algorithms and their robustness to various image transformations. The algorithms&#x2019; performance is tested on two image sets in the different spectra: digital X-Ray images and images taken in the visible spectrum. Each dataset captures complex scenes with many objects and partial occlusions. Geometrical transformations (rotation, shearing, scaling), linear color transformations, Gaussian blur are applied to the images. Then the detection and description algorithms are tested on the original and transformed images. The repeatability and number of corresponding points are calculated to assess detection algorithms. The ratio of correctly matched descriptors together with the ratio of the distances between the query descriptor, the nearest descriptor, and the second matched descriptor is computed to evaluate descriptors&#x2019; quality. The algorithms showed different behavior on different spectra. SURF demonstrated to be the best X-ray keypoint detector and for the visible spectrum, it shares first place with AKAZE detector. SIFT is the best descriptor in both spectra. The strong and weak points of each algorithm are discussed in the paper.

Multi-Level Feature Aggregation-Based Joint Keypoint Detection and Description

Image keypoint detection and description is a popular method to find pixel-level connections between images, which is a basic and critical step in many computer vision tasks. The existing methods are far from optimal in terms of keypoint positioning accuracy and generation of robust and discriminative descriptors. This paper proposes a new end-to-end self-supervised training deep learning network. The network uses a backbone feature encoder to extract multi-level feature maps, then performs joint image keypoint detection and description in a forward pass. On the one hand, in order to enhance the localization accuracy of keypoints and restore the local shape structure, the detector detects keypoints on feature maps of the same resolution as the original image. On the other hand, in order to enhance the ability to percept local shape details, the network utilizes multi-level features to generate robust feature descriptors with rich local shape information. A detailed comparison with traditional feature-based methods Scale Invariant Feature Transform (SIFT), Speeded Up Robust Features (SURF) and deep learning methods on HPatches proves the effectiveness and robustness of the method proposed in this paper.

Can keypoint descriptors be used in outdoor image classification?

In this paper, we approach a Content Based Image Retrieval problem using keypoints and texture. We employ an algorithm that allows computing a fixed number of keypoints for all the images in the dataset, by adapting a certain parameter of the keypoints computation method. The SIFT (Scale Invariant Feature Transform) and SURF (Speeded Up Robust Features) algorithms yield the required keypoints. We also used texture features to improve the retrieval results. The texture features are computed using Local Binary Patterns (LBP), Gray Level Co-occurrence Matrices (GLCM), Dual Tree Complex Wavelet Transform (DTCWT), and Gabor filters. The methods were tested on our own dataset, ODIDS which contains outdoor images, mainly acquired in the city of Iaşi, Romania. We performed location identification with good results.

Image processing and machine learning based cavings characterization and classification

Wellbore failure is the most headache and common problem in drilling engineering. Wireline loggings like caliper log and image log are usually used to identify wellbore failure. However, it is impossible to discern the failure types and take steps in real-time using the post-drilling methods. During the drilling process, various cavings will circulate to the ground with drilling fluid when wellbore instability occurs. These cavings are the foremost indicator of wellbore deterioration and have a potential link to the mechanism of borehole failure. Unfortunately, the use of cavings to understand borehole instability and its mechanism requires correct descriptions of caving which are inclined to be qualitative and fragmentary in previous studies, as well as proper interpretation which entails a wealth of field experience and comprehensive understanding of drilling, geo-mechanics and geology. This paper proposes a new method for caving characterization and a caving interpretation workflow based on image processing and machine learning. Firstly, we thoroughly analyze all modes of borehole failures and the corresponding cavings they generate, and make clear the significant characteristics to distinguish different types of cavings. Then the morphologies of caving contour and surface are quantitatively characterized using Speeded-Up Robust Features (SURF) algorithm and 2D- Discrete Wavelet Transform (DWT), respectively. These two categories of characteristics are used as the input of Light Gradient Boosting Machine (LightGBM)-based classification model, and the mode of borehole failure that generates the caving can be inferred from the classification results. We demonstrate the application of this interpretation workflow utilizing the cavings from the T oilfield, Xinjiang, China. The result shows that the average accuracy of classification is 90.9 %, which reflects that the workflow has a satisfactory performance in caving interpretation as well as that the features have high representativeness and robustness for caving characterization. This method opens up a new path to the early warning of borehole instability and real-time diagnosis of instability mechanism, thus corresponding remedies can be taken in time to avoid serious downhole accidents. • The differences of morphology between various types of cavings are summarized into two aspects: contour and surface. • Contour morphology is quantitatively characterized based on the SURF algorithm. • 2D-DWT is applied to extract the characteristics of surface morphology at multi scales. • A caving interpretation model that links the morphology of caving to the mode of borehole instability is built.

Multi-scale super-pixels-based passive forensics for copy-move forgery using local characteristics

Copy-move forgery is the most common kind of tampering technique for digital images. This paper presents a novel hybrid approach, which uses the Speeded-Up Robust Features (SURF) point and characteristics of Local Feature Regions (LFRs) matching. First, the multi-scale super-pixels algorithm adaptively divides the suspect image into irregular blocks according to the texture level of host images. Then, the improved SURF detector is adopted in extracting feature points from each super-pixel and the feature point threshold is related to the entropy of each super-pixel block. Next, LFRs are defined, and a robust feature descriptor is extracted from each LFR as a vector field. Last, the matching LFRs are found by using Euclidean locality sensitive hashing; the removal of falsely matched pairs is realised by using the Random Sample Consensus (RANSAC) algorithm. Comparing with the leading-edge block-matching methods and point-based methods, our method can produce far better detection results.

Visual Vocabulary Based Photovoltaic Health Monitoring System Using Infrared Thermography

Photovoltaic (PV) systems have gained global acceptance in terms of green, replenishable energy resources to meet energy demand with no emissions. However, PV systems are susceptible to operational and environmental stresses. Moreover, PV panels monitoring is necessary to keep their performance and efficiency intact due to their lack of supervisory control. Therefore, this study monitors PV panels based on health into three sub-classes: healthy, hotspot, and faulty through infrared thermography. First, Thermographs key points are selected using an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$8\times 8$ </tex-math></inline-formula> uniform pixel grid, and speed-up robust features (SURF) are extracted from grid intersection points. Afterward, due to its simplicity, the k-mean clustering algorithm creates single-level clusters based on actual observations similarities and similar observations closeness within-cluster and dissimilarity to other clusters observations are used to transform features into visual words. Finally, shallow classifiers are utilized because of low training time and high prediction speed. After extensive testing and compressive analysis, the proposed approach was found economical, fast, and showed high testing accuracy of 97% through a multi-class shallow classifier (support vector machine) with low computational complexity and less storage size. Thus, this approach can monitor megawatt PV systems with high accuracy and keep performance and emissions mitigation potential high while lowering payback time.

KAZE-SAR: SAR Image Registration Using KAZE Detector and Modified SURF Descriptor for Tackling Speckle Noise

Synthetic aperture radar (SAR) image registration is still a challenging task in remote sensing. The scale-invariant feature transform (SIFT) method and its extensions are most widely used feature detectors in SAR images. However, due to the presence of nonlinear speckle noise, some wrong features are chosen which directly influence the feature matching process. In this article, for the first time, the KAZE algorithm with a modified version of speeded-up robust features (SURF) descriptor is used for SAR image registration. It uses discrete stochastic second order nonlinear partial differential equations (PDEs) to model the SAR image’s edge structure. KAZE uses nonlinear diffusion filtering to build up the scale levels of the SIFT descriptor. It preserves the edges while simultaneously decreasing the speckle noise in smoothing the image. Therefore, it captures the exact location of features in down-sampling and filtering stages of SIFT. Experimental results show that our proposed method has increased the localization accuracy and distinctiveness of feature extraction through the use of KAZE detector. Our method is also able to handle challenging cases in different image sources, different view angles, different image times, complex affine mapping, presence of blurring, and noise.

Gauss Gradient and SURF Features for Landmine Detection from GPR Images

Recently, ground-penetrating radar (GPR) has been extended as a well-known area to investigate the subsurface objects. However, its output has a low resolution, and it needs more processing for more interpretation. This paper presents two algorithms for landmine detection from GPR images. The first algorithm depends on a multi-scale technique. A Gaussian kernel with a particular scale is convolved with the image, and after that, two gradients are estimated; horizontal and vertical gradients. Then, histogram and cumulative histogram are estimated for the overall gradient image. The bin values on the cumulative histogram are used for discrimination between images with and without landmines. Moreover, a neural classifier is used to classify images with cumulative histograms as feature vectors. The second algorithm is based on scale-space analysis with the number of speeded-up robust feature (SURF) points as the key parameter for classification. In addition, this paper presents a framework for size reduction of GPR images based on decimation for efficient storage. The further classification steps can be performed on images after interpolation. The sensitivity of classification accuracy to the interpolation process is studied in detail.

Farmland Aerial Images Fast-Stitching Method and Application Based on Improved SIFT Algorithm

The main black land conservation measure in China is the straw return to the fields. The processing of high-resolution images collected by aerial photography of UAVs through image stitching technology can provide image information for achieving fast and accurate detection of straw cover over large areas. The classical SIFT algorithm has many drawbacks, such as high dimensionality of feature descriptors, high computational effort, and low matching efficiency. To solve the problems above, this study proposes an improved algorithm. First, the method down sampled the high-resolution images before detecting the features to reduce the number of feature points and improve the efficiency of feature detection. Then, matching among feature points is achieved by gradient normalization-based feature descriptors to improve the matching accuracy. Next, the Progressive Sample Consistency algorithm eliminates the mismatch points and optimizes the transformation model. Finally, the images are fused with optimal stitching combined with fade-in and fade-out to achieve high-quality stitching. The comparative experimental results show that compared with the traditional SIFT and the speed-up robust feature algorithms, the algorithm has the advantage of the speed and good robustness to angle rotation, and makes full use of the texture information and the detail information, so it has higher accuracy. Compared with the traditional methods, the panoramic stitching image quality herein is excellent and can be applied to subsequent straw cover detection, the straw cover error is ≤3%, meeting the demand for large-area straw cover detection. Overall, the method proposed herein achieves an ideal balance between accuracy and efficiency; and outperforms other widely used and superior methods.

LUNG CANCER DETECTION TECHNIQUE BASED ON SURF DESCRIPTOR AND KNN ALGORITHMS

In this century, lung cancer is undoubtedly one of the major serious health problems, and one of the leading causes of death for women and men worldwide. Despite advances in treating lung cancer with unprecedented products of pharmaceutical and technological advances, mortality and morbidity rates remain a major challenge for oncologists and cancer biologists. Thus, there is an urgent need to provide early, accurate, and effective diagnostic techniques to improve the survival rate and reduce morbidity and mortality related to lung cancer patients. Therefore, in this paper, an effective lung cancer screening technique is proposed for the early detection of risk factors for lung cancer. In this proposed technique, the powerful acceleration feature Speeded up robust feature (SURF) was used to extract the features. One of the machine learning methods was used to detect cancer by relying on the k nearest neighbor (KNN ) method, where the experimental results show an effective way to discover SURF features and tumor detection by relying on neighborhoods and calculating the distance using KNN. As a result, a high system sensitivity performance success rate of 96% and a system accuracy of 99% has been achieved.

A Text Tracking Method using Maximally Stable Extremal Regions and Speeded up Robust Features

Content-based image retrieval is an active research area because of vast applications of image and video collections. Text embedded in video can provide significant contribution in identifying the contents of the multimedia data and facilitate the process of video indexing, retrieval and analysis. Text tracking is a vital part of text extraction process. It can speed up the text extraction process and also improves the text localization accuracy for videos. This paper proposes a novel approach for text tracking in videos. This method experimentally proposes Maximally Stable Extremal Regions (MSER) and Speeded Up Robust Features (SURF) descriptors for text tracking in videos. MSER is used for interest point detection as Extremal regions are affine invariant, so it is a suitable option for text tracking which can undergo scale, rotation and translation changes. SURF is chosen as a feature descriptor because of its scale and rotation invariance, speed and robustness. Proposed technique is testified on two datasets; the first is designed to test the tracking methodology as part of this research and second is publicly available Youtube Video Text(YVT) dataset. Succeeding analysis on diverse datasets endures verification to the fact that the projected technique demonstrates visible improvements to text tracking in videos

A Novel Hybrid Approach Based on Deep CNN to Detect Glaucoma Using Fundus Imaging

Glaucoma is one of the eye diseases stimulated by the fluid pressure that increases in the eyes, damaging the optic nerves and causing partial or complete vision loss. As Glaucoma appears in later stages and it is a slow disease, detailed screening and detection of the retinal images is required to avoid vision forfeiture. This study aims to detect glaucoma at early stages with the help of deep learning-based feature extraction. Retinal fundus images are utilized for the training and testing of our proposed model. In the first step, images are pre-processed, before the region of interest (ROI) is extracted employing segmentation. Then, features of the optic disc (OD) are extracted from the images containing optic cup (OC) utilizing the hybrid features descriptors, i.e., convolutional neural network (CNN), local binary patterns (LBP), histogram of oriented gradients (HOG), and speeded up robust features (SURF). Moreover, low-level features are extracted using HOG, whereas texture features are extracted using the LBP and SURF descriptors. Furthermore, high-level features are computed using CNN. Additionally, we have employed a feature selection and ranking technique, i.e., the MR-MR method, to select the most representative features. In the end, multi-class classifiers, i.e., support vector machine (SVM), random forest (RF), and K-nearest neighbor (KNN), are employed for the classification of fundus images as healthy or diseased. To assess the performance of the proposed system, various experiments have been performed using combinations of the aforementioned algorithms that show the proposed model based on the RF algorithm with HOG, CNN, LBP, and SURF feature descriptors, providing ≤99% accuracy on benchmark datasets and 98.8% on k-fold cross-validation for the early detection of glaucoma.

Thermal Drift Correction for Laboratory Nano Computed Tomography via Outlier Elimination and Feature Point Adjustment.

Thermal drift of nano-computed tomography (CT) adversely affects the accurate reconstruction of objects. However, feature-based reference scan correction methods are sometimes unstable for images with similar texture and low contrast. In this study, based on the geometric position of features and the structural similarity (SSIM) of projections, a rough-to-refined rigid alignment method is proposed to align the projection. Using the proposed method, the thermal drift artifacts in reconstructed slices are reduced. Firstly, the initial features are obtained by speeded up robust features (SURF). Then, the outliers are roughly eliminated by the geometric position of global features. The features are refined by the SSIM between the main and reference projections. Subsequently, the SSIM between the neighborhood images of features are used to relocate the features. Finally, the new features are used to align the projections. The two-dimensional (2D) transmission imaging experiments reveal that the proposed method provides more accurate and robust results than the random sample consensus (RANSAC) and locality preserving matching (LPM) methods. For three-dimensional (3D) imaging correction, the proposed method is compared with the commonly used enhanced correlation coefficient (ECC) method and single-step discrete Fourier transform (DFT) algorithm. The results reveal that proposed method can retain the details more faithfully.

A Demodulation Algorithm for Periodically In-Plane Vibrating MEMS Based on a Stroboscopic Micro-Visual System.

A high-frequency short-pulsed stroboscopic micro-visual system was employed to capture the transient image sequences of a periodically in-plane working micro-electro-mechanical system (MEMS) devices. To demodulate the motion parameters of the devices from the images, we developed the feature point matching (FPM) algorithm based on Speeded-Up Robust Features (SURF). A MEMS gyroscope, vibrating at a frequency of 8.189 kHz, was used as a testing sample to evaluate the performance of the proposed algorithm. Within the same processing time, the SURF-based FPM method demodulated the velocity of the in-plane motion with a precision of 10−5 pixels of the image, which was two orders of magnitude higher than the template-matching and frame-difference algorithms.

MRI brain tumor image classification with support vector machine

Digital image processing is becoming a developing research arena in medical disciplines for various pathological operations for detection of brain tumor and classification and also for examining and testing critical parts of the human body using microscopic images. In the proposed work segmentation is done by hybridizing the traditional k-means algorithm with SGHO (swarm -based grass hopper optimization algorithm). The SURF (speeded up robust feature) algorithm has been applied to extract features of the brain tumor images and SGHO based technique is used for selecting the features. In the last step svm classifier is applied for the classification of the tumor images. For performing all the steps of the proposed system, publicly accessible Contrast- Enhanced MRI dataset is utilized. The values for accuracy, precision and recall are 99.24%, 95.83%, and 95.30 % respectively. In terms of performance parameters, the results shows that the efficiency of the system is better when compared with earlier works.

Machine Learning-Based Approaches for Tomato Pest Classification

Insect pests are posing a significant threat to agricultural production. They live in different places like fruits, vegetables, flowers, and grains. It impacts plant growth and causes damage to crop yields. We presented an automatic detection and classification of tomato pests using image processing with machine learning-based approaches. In our work, we considered texture features of pest images extracted by feature extractors like gray level co-occurrence matrix (GLCM), local binary pattern (LBP), histogram of oriented gradient (HOG), and speeded up robust features (SURF). The three standard classification methods, including Support Vector Machine (SVM), k-Nearest Neighbour (k-NN), and Decision Tree (DT), are used to classify features of tomato pest by feature extractors mentioned above. The three classifiers have undergone a comprehensive analysis to present which classifier yields the best accuracy, and the value of best accuracy obtained using the feature extraction method. The experiment results showed that the SVM classifier's precision using the feature extracted by LBP achieves the highest value of 81.02%. MATLAB software used for feature extraction and WEKA graphical user interface for classification.

A novel image‐orientation feature extraction method for partial discharges

Partial discharge identification is of great significance in monitoring power equipment. Although substantial work has been done in partial discharge pattern recognition, most studies have been performed in noise-free environments. To improve the accuracy of partial discharge pattern recognition, this paper introduces a partial discharge image feature extraction method based on upright speeded-up robust features. Each group of phase-resolved pulse sequence data is converted into grayscale images to construct a sample library of partial discharge defects. Upright speeded-up robust features is used to extract features in grayscale images. The improved support vector machine is used to achieve the best classification of partial discharge sources. The sensitivity of local features to different phase resolutions and different image resolutions of partial discharge phase-resolved pulse sequence patterns was investigated, and the optimal phase resolution and image resolution were used to construct phase-resolved pulse sequence patterns. In addition, the recognition effect of this feature extraction method under different noise conditions was also analysed. The results prove that upright speeded-up robust features can effectively deal with noisy data. The results of this research can provide references for the storage setting and recognition of three-dimensional images of on-site partial discharge data.

An Imperceptible Watermarking Scheme for Medical Image Tamper Detection

One of the important issues in telemedicine field refers to an advanced secure communication. Digital image watermarking is an ideal solution since it protects the electronic patient information’s from unauthorized access. This paper presents a novel blind fragile-based image watermarking scheme in spatial domain that merges Speed Up Robust Features (SURF) descriptor with the well-known Weber Descriptors (WDs) and Arnold algorithm. It provides a good way for enhancing the image quality and time complexity for medical data integrity. Firstly, the watermark image is shuffled using Arnold chaotic map. Secondly, the SURF technique is practiced to Region of Interest (ROI) of the medical image and then the blocks around the SURF points are selected to insert the watermark. Finally, the watermark is encrusted and extracted using WDs. Experimental results show good image fidelity with the shortest execution time to ensure medical images integrity.