Sliding Window Search Research Articles

Vision-Based Algorithm for Precise Traffic Sign and Lane Line Matching in Multi-Lane Scenarios

With the rapid development of intelligent transportation systems, lane detection and traffic sign recognition have become critical technologies for achieving full autonomous driving. These technologies offer crucial real-time insights into road conditions, with their precision and resilience being paramount to the safety and dependability of autonomous vehicles. This paper introduces an innovative method for detecting and recognizing multi-lane lines and intersection stop lines using computer vision technology, which is integrated with traffic signs. In the image preprocessing phase, the Sobel edge detection algorithm and weighted filtering are employed to eliminate noise and interference information in the image. For multi-lane lines and intersection stop lines, detection and recognition are implemented using a multi-directional and unilateral sliding window search, as well as polynomial fitting methods, from a bird’s-eye view. This approach enables the determination of both the lateral and longitudinal positioning on the current road, as well as the sequencing of the lane number for each lane. This paper utilizes convolutional neural networks to recognize multi-lane traffic signs. The required dataset of multi-lane traffic signs is created following specific experimental parameters, and the YOLO single-stage target detection algorithm is used for training the weights. In consideration of the impact of inadequate lighting conditions, the V channel within the HSV color space is employed to assess the intensity of light, and the SSR algorithm is utilized to process images that fail to meet the threshold criteria. In the detection and recognition stage, each lane sign on the traffic signal is identified and then matched with the corresponding lane on the ground. Finally, a visual module joint experiment is conducted to verify the effectiveness of the algorithm.

A Robust Registration Method for Multi-view SAR Images based on Best Buddy Similarity

Abstract. Due to the influence of imaging angle and terrain undulation, multi-view synthetic aperture radar (SAR) images are difficult to be directly registered by traditional methods. Although feature matching solves the issue of image rotation and maintains scale invariance, these methods often lead to non-uniformity of interest points and may not achieve subpixel accuracy. The traditional template matching method makes it difficult to generate correct matches for multi-view SAR oblique images. In this paper, a multi-view SAR image template matching method based on Best Buddy Similarity (BBS) is proposed to solve the traditional methods' problem. Firstly, the initial correspondences between images are established according to the Range-Doppler model of SAR images. Secondly, a sliding window search is performed on the established correspondence, the BBS is calculated, and the subpixel locations of the peaks on the similarity map are estimated to achieve a fine match. In the calculation process of BBS, the SAR-ROEWA operator is used to suppress the speckle noise of SAR images. The experiment demonstrated that SAR-BBS can accurately match SAR images with large rotation angle. The peak value on the search window is significant. The registration accuracy of SAR-BBS outperforms the other state-of-the-art methods.

Reliability analysis of a two-dimensional linear consecutive-(r,s)-out-of-(m,n): F repairable system

Most of the existing studies on the two-dimensional consecutive system have focused on the non-repairable case. Nevertheless, considering the repairable case of components in engineering, it is necessary to study the reliability of the two-dimensional consecutive repairable system. Owing to the more complex spatial structure of the two-dimensional consecutive system, the analysis methods of the one-dimensional consecutive repairable system are not applicable to the two-dimensional case. In this work, we give a general method to analyze the reliability of a two-dimensional linear consecutive-(r,s)-out-of-(m,n): F repairable system. Firstly, to calculate the number of different cases of the system in a certain state, a key issue in the analysis of the consecutive repairable system, we propose the sliding window search method and inverse search method, whose computational complexities are compared and correctness is verified. Next, under the assumption that the repair time follows a general distribution, we analyze the reliability of the two-dimensional linear consecutive-(r,s)-out-of-(m,n): F repairable system using the supplementary variable method. Finally, we use an example of a wireless sensor network to analyze the effects of system parameters on system reliability indices, such as steady-state availability, steady-state failure frequency, and MTTFF.

CURTAINs for your sliding window: Constructing unobserved regions by transforming adjacent intervals.

We propose a new model independent technique for constructing background data templates for use in searches for new physics processes at the LHC. This method, called Curtains, uses invertible neural networks to parameterise the distribution of side band data as a function of the resonant observable. The network learns a transformation to map any data point from its value of the resonant observable to another chosen value. Using Curtains, a template for the background data in the signal window is constructed by mapping the data from the side-bands into the signal region. We perform anomaly detection using the Curtains background template to enhance the sensitivity to new physics in a bump hunt. We demonstrate its performance in a sliding window search across a wide range of mass values. Using the LHC Olympics dataset, we demonstrate that Curtains matches the performance of other leading approaches which aim to improve the sensitivity of bump hunts, can be trained on a much smaller range of the invariant mass, and is fully data driven.

Scope2Screen: Focus+Context Techniques for Pathology Tumor Assessment in Multivariate Image Data.

Inspection of tissues using a light microscope is the primary method of diagnosing many diseases, notably cancer. Highly multiplexed tissue imaging builds on this foundation, enabling the collection of up to 60 channels of molecular information plus cell and tissue morphology using antibody staining. This provides unique insight into disease biology and promises to help with the design of patient-specific therapies. However, a substantial gap remains with respect to visualizing the resulting multivariate image data and effectively supporting pathology workflows in digital environments on screen. We, therefore, developed Scope2Screen, a scalable software system for focus+context exploration and annotation of whole-slide, high-plex, tissue images. Our approach scales to analyzing 100GB images of 109 or more pixels per channel, containing millions of individual cells. A multidisciplinary team of visualization experts, microscopists, and pathologists identified key image exploration and annotation tasks involving finding, magnifying, quantifying, and organizing regions of interest (ROIs) in an intuitive and cohesive manner. Building on a scope-to-screen metaphor, we present interactive lensing techniques that operate at single-cell and tissue levels. Lenses are equipped with task-specific functionality and descriptive statistics, making it possible to analyze image features, cell types, and spatial arrangements (neighborhoods) across image channels and scales. A fast sliding-window search guides users to regions similar to those under the lens; these regions can be analyzed and considered either separately or as part of a larger image collection. A novel snapshot method enables linked lens configurations and image statistics to be saved, restored, and shared with these regions. We validate our designs with domain experts and apply Scope2Screen in two case studies involving lung and colorectal cancers to discover cancer-relevant image features.

Moving-Vehicle Identification Based on Hierarchical Detection Algorithm

The vehicle detection method plays an important role in the driver assistance system. Therefore, it is very important to improve the real-time performance of the detection algorithm. Nowadays, the most popular method is the scanning method based on sliding window search, which detects the vehicle from the image to be detected. However, the existing sliding window detection algorithm has many drawbacks, such as large calculation amount and poor real-time performance, and it is impossible to detect the target vehicle in real time during the motion process. Therefore, this paper proposes an improved hierarchical sliding window detection algorithm to detect moving vehicles in real time. By extracting the region of interest, the region of interest is layered, the maximum and minimum values of the detection window in each layer are set, the flashing frame generated by the layering is eliminated by the delay processing method, and a method suitable for the motion is obtained: the real-time detection algorithm of the vehicle, that is, the hierarchical sliding window detection algorithm. The experiments show that the more layers are divided, the more time is needed, and when the number of detection layers is greater than 7, the time change rate increases significantly. As the number of layers decreases, the detection accuracy rate also decreases, resulting in the phenomenon of a false positive. Therefore, it is determined to meet the requirements of real time and accuracy when the image is divided into 7 layers. It can be seen from the experiment that when the images to be detected are divided into 7 layers and the maximum and minimum values of detection windows are 30 × 30 and 250 × 250, respectively, the number of sub-windows generated is one thirty-seventh of the original sliding window detection algorithm, and the execution time is only one-third of the original sliding window detection algorithm. This shows that the hierarchical sliding window detection algorithm has better real-time performance than the original sliding window detection algorithm.

Advanced lane detection technique for structural highway based on computer vision algorithm

Lane departure detection plays a vital role in the Advanced Driver assistive systems and it improves the vehicle's active safe driving. A wholesome lane detection method which is based on computer vision techniques, is introduced. The lane boundaries and its radius of curvatures and lane direction were detected from a stream of videos. This video footage was recorded from a camera mounted on the top of a vehicle. We have corrected the camera distortion in the input frame. HLS thresholding and sobel thresholding techniques are applied to the undistorted image for getting focus on the lane lines in the binary image. Then the resulted frame is warped to the bird’s eye by applying the perspective transform. The respective lane line pixels are identified using sliding window search and then left and right lane lines are identified by fitting second-degree polynomials. The lane curvature and deviation from the lane centre are also computed after the identification of the lane. The identified lane boundaries are warped back onto the input image and the radius of lane curvature and vehicle position is calculated and displayed. Hence this technique is enforced using python programming language and for processing the images open CV is utilized. The obtained result illustrates how the proposed method accurately detects the lane line in different lightning conditions.

Object Pose Estimation in Monocular Image Using Modified FDCM

In this paper, a new method for object detection and pose estimation in a monocular image is proposed based on FDCM method. it can detect object with high speed running time, even if the object was under the partial occlusion or in bad illumination. In addition, It requires only single template without any training process. The Modied FDCM based on FDCM with improvments, the LSD method was used in MFDCM instead of the line tting method, besides the integral distance transform was replaced with a distance transform image, and using an angular Voronoi diagram. In addition, the search process depends on Line segments based search instead of the sliding window search in FDCM. The MFDCM was evaluated by comparing it with FDCM in dierent scenarios and with other four methods: COF, HALCON, LINE2D, and BOLD using D-textureless dataset. The comparison results show that MFDCM was at least 14 times faster than FDCM in tested scenarios. Furthermore, it has the highest correct detection rate among all tested method with small advantage from COF and BLOD methods, while it was a little slower than LINE2D which was the fasted method among compared methods. The results proves that MFDCM able to detect and pose estimation of the objects in the clear or clustered background from a monocular image with high speed running time, even if the object was under the partial occlusion which makes it robust and reliable for real-time applications.

The Power Line Inspection Software (PoLIS): A versatile system for automating power line inspection

A large amount of data, provided in the form of video data, is acquired during manned inspections flights of electric power lines. This data is analyzed by expert human inspectors to detect faults in the power lines infrastructure and prepare the inspection reports. This process is extremely time consuming, very expensive and prone to human error. In this paper, we present PoLIS: the Power Line Inspection Software, which has been developed with the objective of assisting the analysis of the data acquired during inspection flights. PoLIS is based on the cooperation between computer vision and machine learning techniques to automatically process video sequences acquired during inspection flights, resulting in a set of representative images per electric tower which we call Key Frames. These representative images can then be used for inspection purposes, leading to a drastic reduction of the human operators’ workload. At the core of the strategy lies an electric tower detector, which is in charge of estimating the location of the towers within the images based on the combination of a sliding window search technique and a supervised classifier. The location of the tower is then tracked using a tracking-by-registration algorithm based on direct methods, estimating the position of the tower in different images. Finally, different criteria are applied for defining whether the image corresponds to a Key Frame image or not. Extensive evaluation of the proposed strategy is conducted using videos acquired during manned helicopter inspections. The videos constituting this database contain several thousand frames representing both medium and high voltage power transmission lines in the infra-red (IR) and visible spectra. The obtained results show that the proposed strategy can reduce the large amount of data present in the inspection videos to a few Key Frames for each tower. It is also demonstrated that the learning-based approach proposed in PoLIS is appropriate for detecting electric towers, a process which is made faster and more robust by coupling it with a tower tracking algorithm. A Graphical User Interface allowing the application of PoLIS to user-provided videos is also presented in this paper, illustrating the whole process and the automated generation of an inspection report.

Research on Real Time ICP Algorithm for Geomagnetic Matching Simulation

The application of geomagnetic field to auxiliary matching navigation is a new direction in the research field of integrated navigation technology. This technology provides a new means for passive location of underwater submersible. The nearest point iterative method is one of the important matching navigation algorithms, but it has some disadvantages such as low real-time performance and slow search speed. Considering the above two shortcomings, we adopt the fixed initial sequence length to improve the sampling structure and deduce the single point iteration formula. Meanwhile, we use sliding window search to narrow the search area and speed up the algorithm, and finally realize the real-time ICP algorithm design. Based on the MATLAB platform, we realized the real-time ICP algorithm geomagnetic matching simulation experiment. The simulation system first used a large area of aeromagnetic anomaly data to build geomagnetic map for matching and positioning experiment. Then the inertial / geomagnetic integrated navigation system is used to experiment on the small area geomagnetic map of self-test, and the simulation results are better.

Object Proposal Generation With Fully Convolutional Networks

Object proposal generation, as a preprocessing technique, has been widely used in current object detection pipelines to guide the search of objects and avoid exhaustive sliding window search across images. Current object proposals are mostly based on low-level image cues, such as edges and saliency. However, objectness is possibly a high-level semantic concept showing whether one region contains objects. This paper presents a framework utilizing fully convolutional networks (FCNs) to produce object proposal positions and bounding box location refinement with Support Vector Machine (SVM) to further improve proposal localization. Experiments on the PASCAL VOC 2007 show that using high-level semantic object proposals obtained by FCN, the object recall can be improved. An improvement in detection mean average precision is also seen when using our proposals in the Fast R-convolutional neural network framework. In addition, we also demonstrate that our method shows stronger robustness when introduced to image perturbations, e.g., blurring, JPEG compression, and salt and pepper noise. Finally, the generalization capability of our model (trained on the PASCAL VOC 2007) is evaluated and validated by testing on PASCAL VOC 2012 validation set, ILSVRC 2013 validation set, and MS COCO 2014 validation set.

Greedy refinement of object proposals via boundary‐aligned minimum bounding box search

Recently developed object detectors rely on automatically generated object proposals, instead of using a dense sliding window search scheme; generating good object proposals has therefore become crucial for improving the computational cost and accuracy of object detection performance. In particular, the shape and location errors of object proposals can be directly propagated to object detection unless some additional processes are adopted to refine the shape and location of bounding boxes. In this study, the authors demonstrate an object proposal refinement algorithm that improves the localisation accuracy and refines the shape of object proposals by searching a boundary‐aligned minimum bounding box. They assume that an object consists of several image regions, and that the optimal object proposal is well aligned with image region boundaries. Based on this assumption, they design novel boundary‐region alignment measures and then propose a greedy refinement method based on the proposed measures. Experiments on the PASCAL VOC 2007 dataset show that the proposed method produces highly well‐localised object proposals and truly improves the quality of object proposals.

Toward Fast and Accurate Vehicle Detection in Aerial Images Using Coupled Region-Based Convolutional Neural Networks

Vehicle detection in aerial images, being an interesting but challenging problem, plays an important role for a wide range of applications. Traditional methods are based on sliding-window search and handcrafted or shallow-learning-based features with heavy computational costs and limited representation power. Recently, deep learning algorithms, especially region-based convolutional neural networks (R-CNNs), have achieved state-of-the-art detection performance in computer vision. However, several challenges limit the applications of R-CNNs in vehicle detection from aerial images: 1) vehicles in large-scale aerial images are relatively small in size, and R-CNNs have poor localization performance with small objects; 2) R-CNNs are particularly designed for detecting the bounding box of the targets without extracting attributes; 3) manual annotation is generally expensive and the available manual annotation of vehicles for training R-CNNs are not sufficient in number. To address these problems, this paper proposes a fast and accurate vehicle detection framework. On one hand, to accurately extract vehicle-like targets, we developed an accurate-vehicle-proposal-network (AVPN) based on hyper feature map which combines hierarchical feature maps that are more accurate for small object detection. On the other hand, we propose a coupled R-CNN method, which combines an AVPN and a vehicle attribute learning network to extract the vehicle's location and attributes simultaneously. For original large-scale aerial images with limited manual annotations, we use cropped image blocks for training with data augmentation to avoid overfitting. Comprehensive evaluations on the public Munich vehicle dataset and the collected vehicle dataset demonstrate the accuracy and effectiveness of the proposed method.

An improved real-time object proposals generation method based on local binary pattern

Generating a group of category-independent proposals of objects in an image within a very short time is an effective approach to accelerate traditional sliding window search, which has been widely used in preprocessing step of object recognition. In this article, we propose a novel object proposals generation method to produce an order set of candidate windows covering most of object instances. With combination of gradient and local binary pattern, our approach achieves better performance than BING in finding occluded objects and objects in dim lighting conditions. In experiments on the challenging PASCAL VOC 2007 data set, we show that our approach is significantly more accurate than BING. In particular, using 2000 proposals, we achieve 97.6% object detection rate and 69.3% mean average best overlap. Moreover, our proposed method is very efficient and takes only about 0.006 s per image on a laptop central processing unit. The detection speed and high accuracy of proposed method mean that it can be applied to recognizing specific objects in robot visions.

Fast pedestrian detection using deformable part model and pyramid layer location

The majority of pedestrian detection approaches use multiscale detection and the sliding window search scheme with high computing complexity. We present a fast pedestrian detection method using the deformable part model and pyramid layer location (PLL). First, the object proposal method is used rather than the traditional sliding window to obtain pedestrian proposal regions. Then, a PLL method is proposed to select the optimal root level in the feature pyramid for each candidate window. On this basis, a single-point calculation scheme is designed to calculate the scores of candidate windows efficiently. Finally, pedestrians can be located from the images. The Institut national de recherche en informatique et en automatique dataset for human detection is used to evaluate the performance of the proposed method. The experimental results demonstrate that the proposed method can reduce the number of feature maps and windows requiring calculation in the detection process. Consequently, the computing cost is significantly reduced, with fewer false positives.

Scalar controlled induction motor drive speed estimation by adaptive sliding window search of the power signal

This work reports an improved estimation reliability wide range rotor speed estimation algorithm for scalar controlled wound rotor induction machines using spectral search of the stator power signal. An adaptive sliding window speed estimation method is proposed that monitors the value of supply frequency to identify the optimum boundaries of the power signal narrowband maximised by a tracked speed dependent power harmonic, while adjusting to nominal slip variation with supply frequency change characteristic of scalar controlled machines. Analytical expressions are derived that allow direct correlation of the surveyed spectral window boundaries to supply frequency. Once identified, the optimal power spectral narrowband enables supply frequency independent speed evaluation. The proposed method is underpinned by a dichotomous search based frequency tracking technique enabling improvement in the attainable estimation rate. The presented technique’s performance and limitations are assessed in steady-state and transient operation real-time experiments on two different scalar controlled 7.5kW machines. The scheme is shown to provide competent real-time speed estimation in a wide operating speed range for lower dynamics transients and steady-state operation.

Vehicle Detection in Aerial Images Based on Region Convolutional Neural Networks and Hard Negative Example Mining.

Detecting vehicles in aerial imagery plays an important role in a wide range of applications. The current vehicle detection methods are mostly based on sliding-window search and handcrafted or shallow-learning-based features, having limited description capability and heavy computational costs. Recently, due to the powerful feature representations, region convolutional neural networks (CNN) based detection methods have achieved state-of-the-art performance in computer vision, especially Faster R-CNN. However, directly using it for vehicle detection in aerial images has many limitations: (1) region proposal network (RPN) in Faster R-CNN has poor performance for accurately locating small-sized vehicles, due to the relatively coarse feature maps; and (2) the classifier after RPN cannot distinguish vehicles and complex backgrounds well. In this study, an improved detection method based on Faster R-CNN is proposed in order to accomplish the two challenges mentioned above. Firstly, to improve the recall, we employ a hyper region proposal network (HRPN) to extract vehicle-like targets with a combination of hierarchical feature maps. Then, we replace the classifier after RPN by a cascade of boosted classifiers to verify the candidate regions, aiming at reducing false detection by negative example mining. We evaluate our method on the Munich vehicle dataset and the collected vehicle dataset, with improvements in accuracy and robustness compared to existing methods.

Bayesian Reinforcement Learning for Multiscale Combinatorial Grouping

Currently, most of the top performing object detectors apply proposal methods to guide the search for objects, in order to avoiding exhaustive sliding window search. As a classical proposal method, Multiscale Combinatorial Grouping (MCG) [1, 2] performs well on the PASCAL VOC dataset, especially for low proposal number. But when it comes to the autonomous driving object scenarios, the result is poor. In our paper, we applied Bayesian model to the proposals generated by MCG [1, 2] to re-rank the candidate bounding boxes using several geometrical features. We evaluated our method on the challenging KITTI dataset, the results shows that the Bayesian model can greatly improve the performance of MCG [1, 2] for better object detection.

Sensorless speed estimation in wound rotor induction machines by spectral search of the stator phase power signal

This paper investigates the feasibility of utilising the stator phase power signal to establish sensorless speed estimation schemes in wound rotor induction machines (WRIMs). To this end, the authors derive and validate closed form expressions defining the possible power signal spectral content containing rotor speed information. Based on these, two novel sensorless speed estimation techniques are proposed: the first enables supply frequency independent speed estimation in WRIM extended slip applications, while the second facilitates estimation in a wide operating speed range for scalar controlled WRIM applications based on a sliding-window search of the power spectrum. The proposed techniques are underpinned by a dichotomous search enabled spectral processing algorithm and shown to deliver an increase in attainable real-time estimation rate without compromising estimation accuracy. The performance of reported schemes is assessed and validated in real-time laboratory experiments.

One Shot Detection with Laplacian Object and Fast Matrix Cosine Similarity

One shot, generic object detection involves searching for a single query object in a larger target image. Relevant approaches have benefited from features that typically model the local similarity patterns. In this paper, we combine local similarity (encoded by local descriptors) with a global context (i.e., a graph structure) of pairwise affinities among the local descriptors, embedding the query descriptors into a low dimensional but discriminatory subspace. Unlike principal components that preserve global structure of feature space, we actually seek a linear approximation to the Laplacian eigenmap that permits us a locality preserving embedding of high dimensional region descriptors. Our second contribution is an accelerated but exact computation of matrix cosine similarity as the decision rule for detection, obviating the computationally expensive sliding window search. We leverage the power of Fourier transform combined with integral image to achieve superior runtime efficiency that allows us to test multiple hypotheses (for pose estimation) within a reasonably short time. Our approach to one shot detection is training-free, and experiments on the standard data sets confirm the efficacy of our model. Besides, low computation cost of the proposed (codebook-free) object detector facilitates rather straightforward query detection in large data sets including movie videos.