Search results

The purpose of this paper is to propose a fault diagnosis method for rolling bearings, in which the fault feature extraction is realized in a two-dimensional domain using scale invariant feature transform (SIFT) algorithm. This method is different from those methods extracting fault feature directly from the traditional one-dimensional domain.

The vibration signal of rolling bearings is first transformed into a two-dimensional image. Then, the SIFT algorithm is applied to the image to extract the scale invariant feature vector which is highly distinctive and insensitive to noises and working condition variation. As the extracted feature vector is high-dimensional, kernel principal component analysis (KPCA) algorithm is utilized to reduce the dimension of the feature vector, and singular value decomposition technique is used to extract the singular values of the reduced feature vector. Finally, these singular values are introduced into a support vector machine (SVM) classifier to realize fault classification.

The experiment results show a high fault classification accuracy based on the proposed method.

The proposed approach for rolling bearing fault diagnosis based on SIFT-KPCA and SVM is highly effective in the experiment. The practical value in engineering application of this method can be researched in the future.

The purpose of this paper is to present weighted Euclidean distance for measuring whether the fitting of projective transformation matrix is more reliable in feature-based image stitching.

The hybrid model of weighted Euclidean distance criterion and intelligent chaotic genetic algorithm (CGA) is established to achieve a more accurate matrix in image stitching. Feature-based image stitching is used in this paper for it can handle non-affine situations. Scale invariant feature transform is applied to extract the key points, and the false points are excluded using random sampling consistency (RANSAC) algorithm.

This work improved GA by combination with chaos's ergodicity, so that it can be applied to search a better solution on the basis of the matrix solved by Levenberg-Marquardt. The addition of an external loop in RANSAC can help obtain more accurate matrix with large probability. Series of experimental results are presented to demonstrate the feasibility and effectiveness of the proposed approaches.

The modified feature-based method proposed in this paper can be easily applied to practice and can obtain a better image stitching performance with a good robustness.

A hybrid model of weighted Euclidean distance criterion and CGA is proposed for optimization of projective transformation matrix in image stitching. The authors introduce chaos theory into GA to modify its search strategy.

This paper aims to evaluate and compare the performance of different computer vision algorithms in the context of visual servoing for augmented robot perception and autonomy.

The authors evaluated and compared three different approaches: a feature-based approach, a hybrid approach and a machine-learning-based approach. To evaluate the performance of the approaches, experiments were conducted in a simulated environment using the PyBullet physics simulator. The experiments included different levels of complexity, including different numbers of distractors, varying lighting conditions and highly varied object geometry.

The experimental results showed that the machine-learning-based approach outperformed the other two approaches in terms of accuracy and robustness. The approach could detect and locate objects in complex scenes with high accuracy, even in the presence of distractors and varying lighting conditions. The hybrid approach showed promising results but was less robust to changes in lighting and object appearance. The feature-based approach performed well in simple scenes but struggled in more complex ones.

This paper sheds light on the superiority of a hybrid algorithm that incorporates a deep neural network in a feature detector for image-based visual servoing, which demonstrates stronger robustness in object detection and location against distractors and lighting conditions.

Mobile robots become more and more important for many potential applications such as navigation and surveillance. The paper proposes an image processing scheme for moving object detection from a mobile robot with a single camera. It especially aims at intruder detection for the security robot on either smooth paved surfaces or uneven ground surfaces.

The core of the proposed scheme is the template matching with basis image reconstruction for the alignment between two consecutive images in the video sequence. The most representative template patches in one image are first automatically selected based on the gradient energies in the patches. The chosen templates then form a basis matrix, and the instances of the templates in the subsequent image are matched by evaluating their reconstruction error from the basis matrix. For the two well-aligned images, a simple and fast temporal difference can thus be applied to identify moving objects from the background.

The proposed template matching can tolerate in rotation (±10°) and (±10°) in scaling. By adding templates with larger rotational angles in the basis matrixes, the proposed method can be further extended for the match of images from severe camera vibrations. Experimental results of video sequences from a non-stationary camera have shown that the proposed scheme can reliably detect moving objects from the scenes with either minor or severe geometric transformation changes. The proposed scheme can achieve a fast processing rate of 32 frames per second for an image of size 160×120.

The basic approaches for moving object detection with a mobile robot are feature-point match and optical flow. They are relatively computational intensive and complicated to implement for real-time applications. The proposed template selection and template matching are very fast and easy to implement. Traditional template matching methods are based on sum of squared differences or normalized cross correlation. They are very sensitive to minor displacement between two images. The proposed new similarity measure is based on the reconstruction error from the test image and its reconstruction from the linear combination of the templates. It is thus robust under rotation and scale changes. It can be well suited for mobile robot surveillance.

Robot localization in dynamic, cluttered environments is a challenging problem because it is impractical to have enough knowledge to be able to accurately model the robot’s environment in such a manner. This study aims to develop a novel probabilistic method equipped with function approximation techniques which is able to appropriately model the data distribution in Markov localization by using the maximum statistical power, thereby making a sensibly accurate estimation of robot’s pose in extremely dynamic, cluttered indoors environments.

The parameter vector of the statistical model is in the form of positions of easily detectable artificial landmarks in omnidirectional images. First, using probabilistic principal component analysis, the most likely set of parameters of the environmental model are extracted from the sensor data set consisting of missing values. Next, we use these parameters to approximate a probability density function, using support vector regression that is able to calculate the robot’s pose vector in each state of the Markov localization. At the end, using this density function, a good approximation of conditional density associated with the observation model is made which leads to a sensibly accurate estimation of robot’s pose in extremely dynamic, cluttered indoors environment.

The authors validate their method in an indoor office environment with 34 unique artificial landmarks. Further, they show that the accuracy remains high, even when they significantly increase the dynamics of the environment. They also show that compared to those appearance-based localization methods that rely on image pixels, the proposed localization strategy is superior in terms of accuracy and speed of convergence to a global minima.

By using easily detectable, and rotation, scale invariant artificial landmarks and the maximum statistical power which is provided through the concept of missing data, the authors have succeeded in determining precise pose updates without requiring too many computational resources to analyze the omnidirectional images. In addition, the proposed approach significantly reduces the risk of getting stuck in a local minimum by eliminating the possibility of having similar states.

The purpose of this paper is to analyze the development of a novel approach for automated terrain mapping a robotic vehicles path tracing.

The approach includes stitching of images, obtained from unmanned aerial vehicle, based on ORB descriptors, into an orthomosaic image and the GPS-coordinates are binded to the corresponding pixels of the map. The obtained image is fed to a neural network MASK R-CNN for detection and classification regions, which are potentially dangerous for robotic vehicles motion. To visualize the obtained map and obstacles on it, the authors propose their own application architecture. Users can any time edit the present areas or add new ones, which are not intended for robotic vehicles traffic. Then the GPS-coordinates of these areas are passed to robotic vehicles and the optimal route is traced based on this data

The developed approach allows revealing impassable regions on terrain map and associating them with GPS-coordinates, whereas these regions can be edited by the user.

The total duration of the algorithm, including the step with Mask R-CNN network on the same dataset of 120 items was 7.5 s.

Creating an orthophotomap from 120 images with image resolution of 470 × 425 px requires less than 6 s on a laptop with moderate computing power, what justifies using such algorithms in the field without any powerful and expensive hardware.

This paper aims to propose image stitching by reduction of full line and taking line image as registration image to solve the problem of automatic optic inspection in PCB detection. In addition, surf registration was introduced for image stitching to improve the accuracy and speed of stitching.

First, image stitching proceeded by method of full line reduction and taking line image as registration image; second, surf registration was introduced based on the traditional PCB image stitching algorithm. Scale space of the image pyramid was adopted for confirming relative future points between stitching image. The registration means of nearest neighbourhood and next neatest neighborhood was selected for feature matching and fused in region of interest to fulfil image stitching.

The improved stitching algorithm with small data size of image, high speed and noncumulative transitive error eliminated displacement deviation and solved the stitching gap caused by uneven illumination, to greatly improve the accuracy and speed of stitching.

The research of this paper can only used for appearance detection and cannot be used for solder joint inspection with circuit detection or invisible solder joint detection; it can identify and mark PCB component defects but cannot classify automatically, thus artificial confirmation and processing is needed.

Based on the traditional image stitching means, this paper proposed full line reduction for image stitching, which reduces processing of data and speeds up image stitching; in addition, surf registration was introduced into the study of PCB stitching algorithm, which greatly improves the accuracy and speed of stitching and solves stitching gap formed by opposite variation trend of image local edge caused by uneven illumination.

The purpose of this paper is to propose that the three‐dimensional information of obstacles should be identified to allow unmanned aerial vehicles (UAVs) to detect and avoid obstacles existing in their flight path.

First, the approximate outline of obstacles was detected using multi‐scale‐oriented patches (MOPS). At the same time, the spatial coordinates of feature points that exist in the internal outline of the obstacles were calculated through the scale‐invariant feature transform (SIFT) algorithm. Finally, the results from MOPS and the results from the SIFT algorithm were merged to show the three‐dimensional information of the obstacles.

As the method proposed in this paper reconstructs only the approximate outline of obstacles, a quick calculation can be done. Moreover, as the outline information is combined through SIFT feature points, detailed three‐dimensional information pertaining to the obstacles can be obtained.

The proposed approach can be used efficiently in GPS‐denied environments such as certain indoor environments.

For the autonomous flight of small UAVs having a payload limit, this paper suggests a means of forming three‐dimensional information about obstacles with images obtained from a monocular camera.

This paper aims to present a new method for identification of some flight modes, including natural and non-standard modes, and extraction of their characteristics, directly from measurements of flight parameters in the time domain.

The Hilbert-Huang transform (HHT), as a novel prevailing tool in the signal analysis field, is used to attain the purpose. The study shows that the HHT has superior potential capabilities to improve the airplane flying quality analysis and to conquer some drawbacks of the classical method in flight dynamics.

The proposed method reveals the existence of some non-standard modes with small damping ratios at non-linear flight regions and obtains their characteristics.

The paper examines only airplane longitudinal dynamics. Further research is needed regarding lateral-directional dynamic modes and coupling effects of the longitudinal and lateral modes.

Application of the proposed method to the flight test data may result in real-time flying quality analysis, especially at the non-linear flight regions.

First, to utilize the empirical mode decomposition (EMD) capabilities in real time, a local-online algorithm is introduced which estimates the signal trend by the Savitzky-Golay sifting process and eliminates it from the signal in the EMD algorithm. Second, based on the local-online EMD algorithm, a systematic method is proposed to identify flight modes from flight parameters in the time domain.

This study has mainly aimed to compare and contrast two completely different image processing algorithms that are very adaptive for detecting prostate cancer using wearable Internet of Things (IoT) devices. Cancer in these modern times is still considered as one of the most dreaded disease, which is continuously pestering the mankind over a past few decades. According to Indian Council of Medical Research, India alone registers about 11.5 lakh cancer related cases every year and closely up to 8 lakh people die with cancer related issues each year. Earlier the incidence of prostate cancer was commonly seen in men aged above 60 years, but a recent study has revealed that this type of cancer has been on rise even in men between the age groups of 35 and 60 years as well. These findings make it even more necessary to prioritize the research on diagnosing the prostate cancer at an early stage, so that the patients can be cured and can lead a normal life.

The research focuses on two types of feature extraction algorithms, namely, scale invariant feature transform (SIFT) and gray level co-occurrence matrix (GLCM) that are commonly used in medical image processing, in an attempt to discover and improve the gap present in the potential detection of prostate cancer in medical IoT. Later the results obtained by these two strategies are classified separately using a machine learning based classification model called multi-class support vector machine (SVM). Owing to the advantage of better tissue discrimination and contrast resolution, magnetic resonance imaging images have been considered for this study. The classification results obtained for both the SIFT as well as GLCM methods are then compared to check, which feature extraction strategy provides the most accurate results for diagnosing the prostate cancer.

The potential of both the models has been evaluated in terms of three aspects, namely, accuracy, sensitivity and specificity. Each model’s result was checked against diversified ranges of training and test data set. It was found that the SIFT-multiclass SVM model achieved a highest performance rate of 99.9451% accuracy, 100% sensitivity and 99% specificity at 40:60 ratio of the training and testing data set.

The SIFT-multi SVM versus GLCM-multi SVM based comparison has been introduced for the first time to perceive the best model to be used for the accurate diagnosis of prostate cancer. The performance of the classification for each of the feature extraction strategies is enumerated in terms of accuracy, sensitivity and specificity.