Vertical Projection Research Articles

Edge Location Method for Multidimensional Image Based on Edge Symmetry Algorithm

The most basic feature of an image is edge, which is the junction of one attribute area and another attribute area in the image. It is the most uncertain place in the image and the place where the image information is most concentrated. The edge of an image contains rich information. So, the edge location plays an important role in image processing, and its positioning method directly affects the image effect. In order to further improve the accuracy of edge location for multidimensional image, an edge location method for multidimensional image based on edge symmetry is proposed. The method first detects and counts the edges of multidimensional image, sets the region of interest, preprocesses the image with the Gauss filter, detects the vertical edges of the filtered image, and superposes the vertical gradient values of each pixel in the vertical direction to obtain candidate image regions. The symmetry axis position of the candidate image region is analyzed, and its symmetry intensity is measured. Then, the symmetry of vertical gradient projection in the candidate image region is analyzed to verify whether the candidate region is a real edge region. The multidimensional pulse coupled neural network (PCNN) model is used to synthesize the real edge region after edge symmetry processing, and the result of edge location of the multidimensional image is obtained. The results show that the method has strong antinoise ability, clear edge contour, and precise location.

Depth-based branching level estimation for bronchoscopic navigation

Bronchoscopists rely on navigation systems during bronchoscopy to reduce the risk of getting lost in the complex bronchial tree-like structure and the homogeneous bronchus lumens. We propose a patient-specific branching level estimation method for bronchoscopic navigation because it is vital to identify the branches being examined in the bronchus tree during examination. We estimate the branching level by integrating the changes in the number of bronchial orifices and the camera motions among the frames. We extract the bronchial orifice regions from a depth image, which is generated using a cycle generative adversarial network (CycleGAN) from real bronchoscopic images. We calculate the number of orifice regions using the vertical and horizontal projection profiles of the depth images and obtain the camera-moving direction using the feature point-based camera motion estimation. The changes in the number of bronchial orifices are combined with the camera-moving direction to estimate the branching level. We used three in vivo and one phantom case to train the CycleGAN model and four in vivo cases to validate the proposed method. We manually created the ground truth of the branching level. The experimental results showed that the proposed method can estimate the branching level with an average accuracy of 87.6%. The processing time per frame was about 61 ms. Experimental results show that it is feasible to estimate the branching level using the number of bronchial orifices and camera-motion estimation from real bronchoscopic images.

Multi-fold binary Darboux transformation and mixed solitons of a three-component Gross-Pitaevskii system in the spinor Bose-Einstein condensate

The Bose-Einstein-condensation applications give rise to the superfluidity in the liquid helium and superconductivity in the metals. In this paper, we work on a three-component Gross-Pitaevskii system, which describes the matter waves in an spin-1 spinor Bose-Einstein condensate. We construct a multi-fold binary Darboux transformation with the zero seed solutions to describe the three vertical spin projection of the spin-1 spinor BEC, which is different from all the existing Darboux-type ones for the same system, and derive three types of the exponential-and-rational mixed soliton solutions associated with two conjugate complex eigenvalues. For such mixed solitons, we give their asymptotic expressions, indicating that they consist of the Ieda-Miyakawa-Wadati (IMW)-polar-state or IMW-ferromagnetic solitons but possess the time-dependent velocities. Asymptotically and graphically, interaction mechanisms between the mixed and exponential solitons are classified in six cases, and we exhibit the inelastic and elastic interactions through calculating the modifications of the polarization matrices and phase shifts for the two interacting solitons. We find that both the IMW-polar-state solitons, including the mixed and exponential solitons, can not alter the other soliton’s intensity distribution during the interaction, while the mixed or exponential soliton in the IMW-ferromagnetic state does.

3D MSSD: A multilayer spatial structure 3D object detection network for mobile LiDAR point clouds

Point cloud-based object detection is vital and essential for many real-world applications, such as autonomous driving and robot vision. The PointPillars model has achieved the efficient detection of objects in front of a vehicle. However, the algorithm does not consider the spatial structures semantic information stored in the three-dimensional point cloud for a given spatial structure, thus leading to missed or false detections for objects with complex spatial structures or singular structures. We propose an approach based on PointPillars, which considers the spatial structure characteristics of 3D point clouds to enhance the detection accuracy. First, based on the specified range of the z-axis coordinates, the entire point cloud scene is divided into several layers so that the point cloud areas in the same height interval form one layer. Data from several layers are obtained. Second, the point clouds of several layers are processed with Pillar Feature Net to obtain several pseudoimages. Each pseudoimage represents the semantic information from the corresponding level of the point cloud. Third, the obtained pseudoimages from each level are merged with the pseudoimages of the entire scene to obtain a feature map with spatial structure characteristics. We apply a Region Proposal Network, and an object detection operator processes the feature map and obtains the result of object detection. Experiments show that the proposed method has a highly accurate detection effect for objects with complex spatial structures. In addition, the proposed method does not erroneously detect objects with similar semantic information after vertical dimension projection.

Characterizing the Validity of the Inverted Pendulum Model for Quiet Standing.

By assuming that the human body rotates primarily around the ankle joint in the sagittal plane, the human body has been modelled as a single inverted pendulum (IP) to simulate the human quiet stance. Despite its popularity, the validity of the IP model has been challenged in many studies. Rather than testing the validity of the IP model as a true or false question, this work proposes a feature to quantify the degree of validity of the IP model. The development of the proposed feature is based on the fact that the IP model predicts that the horizontal acceleration of COM is proportional to the COP error which is defined as the difference between the center of pressure (COP) and the vertical projection of the center of mass (COM). Since the horizontal components of the acceleration of COM and the ground reaction force (GRF) are always proportional, the proposed feature is the correlation coefficient between the anterior-posterior (AP) components of GRF and the COP error. The efficacy of the proposed feature is demonstrated by comparing its differences for individuals in two age groups (18–24 and 65–73 years) in quiet standing. The experimental results show that the IP model is more suited for predicting the motion of the older group than the younger group. Our results also show that the proposed feature is more sensitive to aging effects than one of the most reliable and accurate COP-based postural stability features.

EXPLOITABLE EDGE ANALYSIS FOR FREE FLOW VEHICLE PLATE LOCALIZATION

In Malaysia, vehicle recognition system (VRS) such as vehicle plate recognition and counting, is rapidly growing and applied in many areas such as to identify vehicle identities for the law enforcement by authorities and electronic toll collection by highway agencies. Uncovering the region of interest in chaotic illumination environment at free flow road makes localizing license plate is critical process of VRS. Available edge vertical projection claimed to be robust to illumination, however, it tends to create false edges and sensitive to noises that can hinder the recognition performance. Thus, this research aims to propose a license plate localization method that based on exploitable edge analysis.comprising four main steps, namely pre-processing, rectangular blob searching, analysis and the vertical rectangular blobs projection. It calculates total and exploit edge information at y axis of the image. The proposed method is then tested on the European number plate datasets i.e. Baza Slika which contains about 167 vehicle images and Ondrej which contains about 97 vehicle images. The experimental results show that the proposed method outperforms the Ondrej method by obtaining accuracy of 95% on Baza Slika dataset and slightly lower by an accuracy of 91% on the Ondrej dataset. Then, the proposed method tested on the Malaysia vehicle dataset namely Tol Sungai Long data set which contains about 584 images of different illumination conditions, i.e. 297 images in the morning, 140 images in the afternoon and 147 images in the night. The proposed method outperforms other approaches with accuracy of 91.24%, 93.57% and 75.51% in the morning, evening and night respectively.

Kurdish Text Segmentation using Projection-Based Approaches

An optical character recognition (OCR) system may be the solution to data entry problems for saving the printed document as a soft copy of them. Therefore, OCR systems are being developed for all languages, and Kurdish is no exception. Kurdish is one of the languages that present special challenges to OCR. The main challenge of Kurdish is that it is mostly cursive. Therefore, a segmentation process must be able to specify the beginning and end of the characters. This step is important for character recognition. This paper presents an algorithm for Kurdish character segmentation. The proposed algorithm uses the projection-based approach concepts to separate lines, words, and characters. The algorithm works through the vertical projection of a word and then identifies the splitting areas of the word characters. Then, a post-processing stage is used to handle the over-segmentation problems that occur in the initial segmentation stage. The proposed method is tested using a data set consisting of images of texts that vary in font size, type, and style of more than 63,000 characters. The experiments show that the proposed algorithm can segment Kurdish words with an average accuracy of 98.6%.

Effect of Positioning on Coelomic Radiography with and without Contrast in the Green Iguana (Iguana iguana)

A prospective study was performed in 12 green iguanas (Iguana iguana) to compare the radiographic appearance of coelomic viscera. Five radiographic projections (views) of the coelom were obtained: dorsoventral, right lateral vertical beam, left lateral vertical beam, right lateral horizontal beam, and left lateral horizontal beam. One week later, the same radiographic projections were obtained after oral administration of barium. In 4 of the 12 green iguanas, an additional vertical beam ventrodorsal view was obtained. In comparing lateral projections, lung, liver, and gonads were best imaged in horizontal beam lateral projections, before distention of the stomach with barium, whereas the gastrointestinal tract including the colon was best visualized in vertical beam lateral projections. The shape and size of the cardiac silhouette was considered equivalent between analogous lateral projections in most green iguanas (10 [83%] of 12). In comparing ventrodorsal and dorsoventral projections, the ventrodorsal projections improved visibility of the cardiac silhouette and lung in 4 (100%) of 4 green iguanas, while reducing cardiac length. Dorsoventral projections were preferred for the evaluation of the gastrointestinal tract. This study demonstrates that radiographic beam orientation and patient position result in obvious radiographic differences of the iguanid coelom and that projections are not interchangeable.

Biomechanics of Single Stair Climb With Implications for Inverted Pendulum Modeling.

Step-by-step (SBS) stair navigation is used by those with movement limitations or lower-limb prosthetics and by humanoid robots. Knowledge of biomechanical parameters for SBS gait, however, is limited. Inverted pendulum (IP) models used to assess dynamic stability have not been applied to SBS gait. This study examined the ability of the linear inverted pendulum (LIP) model and a closed-form, variable-height inverted pendulum (VHIP) model to predict capture-point (CP) stability in healthy adults executing a single stair climb. A second goal was to provide baseline kinematic and kinetic data for SBS gait. Twenty young adults executed a single step onto stairs of two heights, while attached marker positions and ground reaction forces were recorded. opensim software determined body kinematics and joint kinetics. Trials were analyzed with LIP and VHIP models, and the predicted CP compared to the actual center-of-pressure (CoP) on the stair. Lower-limb joint moments were larger than those reported for step-over-step (SOS) stair gait. Leading knee rather than trailing ankle was dominant. Center-of-mass (CoM) velocity peaked at push-off. The VHIP model accounted for only slightly more than half of the forward progression of the vertical projection of the CoM and was not better than LIP predictions. This suggests that IP models are limited in modeling SBS gait, likely due to large hip and knee moments. The results from this study may also provide target values and strategies to aid design of lower-limb prostheses and powered exoskeletons.

Dimension distortion by right coset projections in the Heisenberg group

We study the family of vertical projections whose fibers are right cosets of horizontal planes in the Heisenberg group, \mathbb{H}^n . We prove lower bounds for Hausdorff dimension distortion of sets under these mappings with respect to the natural quotient metric, which we show behaves like the Euclidean metric in this context. Our bounds are sharp in a large part of the dimension range, and we give conjectural sharp lower bounds for the remaining range. Our approach also lets us improve the known almost sure lower bound for the standard family of vertical projections in \mathbb{H}^n for n \geq 2 .

Analysis of Extraction Algorithm for Visual Navigation of Farm Robots Based on Dark Primary Colors

In order to quickly extract the visual navigation line of farmland robot, an extraction algorithm for dark primary agricultural machinery is proposed. The application of dark primary color principle in new farmland is made clearer by gray scale method, and the soil and crops are obviously separated, and the image processing technology of visual navigation line image of farmland is realized. In binary filtering of gray scale images, the maximum interclass variance method and morphological method are used respectively. The researchers use vertical projection method and least square method to the farmland interval extracted by navigation line. The farmland that needs the guide line image will be accurately located. It is found that the visual navigation extraction algorithm of farmland robot is widely used in the image extraction of navigation lines of various farmland roads and scenes compared with the traditional gray scale algorithm. Image processing has the advantages of clearer image processing.

Research on track fastener positioning method based on local unidirectional template matching.

Commonly used fastener positioning methods include pixel statistics (PS) method and template matching (TM) method. For the PS method, it is difficult to judge the image segmentation threshold due to the complex background of the track. For the TM method, the search in both directions of the global is easily affected by complex background, as a result, the locating accuracy of fasteners is low. To solve the above problems, this paper combines the PS method with the TM method and proposes a new fastener positioning method called local unidirectional template matching (LUTM). First, the rail positioning is achieved by the PS method based on the gray-scale vertical projection. Then, based on the prior knowledge, the image of the rail and the surrounding area of the rail is obtained which is referred to as the 1-shaped rail image; then, the 1-shaped rail image and the produced offline symmetrical fastener template is pre-processed. Finally, the symmetrical fastener template image is searched from top to bottom along the rail and the correlation is calculated to realize the fastener positioning. Experiments have proved that the method in this paper can effectively realize the accurate locating of the fastener for ballastless track and ballasted track at the same time.

Image Processing Technology for Edge Detection Based on Vision and Raspberry Pi

Rice (Oryza sativa L) is the main food crop in Indonesia because most of Indonesia’s population consumes rice as a staple food. In this paper, we present a system that functions to detect weeds in rice plants using imaging analysis with a camera device and a Raspberry Pi. The Raspberry Pi got a picture of the field from camera that connected through USB connection and send it to the IoT server. The method used in the analysis is to combine the vertical projection method and the linear scanning method. The results showed that our system was very effective in detecting edges in rice plants.

Coherent artifact and time-dependent polarization in amplified ultrafast erbium-doped fibre lasers

Mode-locked erbium-doped fibre lasers are ultrashort pulsed sources widely studied due to their versatility and multiple applications in the near infrared range. Here we present the experimental study of the emission of a passive mode-locked erbium-doped fibre laser with an amplification stage outside the cavity by means of Frequency Resolved Optical Gating (FROG) and spectral interferometry. Due to shot-to-shot instabilities, the FROG traces can be understood as the combination of two different traces, corresponding to the coherent artifact and the average pulse characteristics. We have modified a Principal Components Generalized Projections Algorithm, in order to make it able to retrieve efficiently both the coherent artifact and the average pulse. In addition, we study the temporal dependence of the polarization, showing that the pulses present time-dependent polarization with a stable spectral relative phase between the horizontal and vertical projections. Up to our knowledge, this is the first experimental study that shows the FROG measurements of unstable pulse trains associated with the coherent artifact and analyses the time-dependent polarization in ultrafast fibre lasers.

Parallel Processing Method of Inertial Aerobics Multisensor Data Fusion

Aerobics is one of the main contents of physical education, which has a positive role in promoting the health of young people. This paper mainly studies the parallel processing method of inertial aerobics multisensor data fusion. In this paper, an aerobics exercise system is designed, which uses digital filter to remove the noise generated in the process of exercise. In this paper, Kalman filter is used to filter the pulse error of accelerometer, and the data structure of unidirectional link is used to achieve the effect of sliding window, which can reduce the memory cost to the greatest extent. In this paper, the region of moving object is determined by horizontal and vertical projection of binary symmetric difference image. At the same time, the optimal feature combination is selected from the reduced features by feature subset selection, and the classification algorithm is used as the evaluation function in the optimization process. Finally, the collected data are tested, analyzed, and sorted out. The experimental data show that, after calibrating the sensor data, the static x-axis and y-axis data are about 0 g, and the z-axis data are about 1 g, which is closer to the real value. The results show that the attitude data collected by the inertial sensor can be stably transmitted to the software of the computer wirelessly for attitude reconstruction, and the recognition of each attitude and parameter has reached a high accuracy.

Sugarcane nodes identification algorithm based on sum of local pixel of minimum points of vertical projection function

Aiming at the difficulty of sugarcane nodes identification and location during automatic cutting of sugarcane seeds, based on machine vision system, this paper proposed a sugarcane nodes identification algorithm based on sum of local pixel of minimum points of vertical projection function. Firstly, according to the color and texture characteristics of yellow sugarcane, the captured RGB image of sugarcane was converted into HSV color image. Then, the S-component image in the HSV image was extracted, and the S-component image was binarized by the Otsu algorithm to obtain the binary image, and then the binary image was processed by morphological closed operation, which can eliminate the noise and fill holes of binary image. Subsequently, the sugarcane area was segmented as the region of interest through the horizontal projection of the binary image, which lowered the amount of calculation while reducing interference. Finally, the vertical projection function of the binary image of the region of interest was established, and the function was continuously derived to obtain the minimum points, and then the position of sugarcane nodes was preliminarily determined by the obtained minimum points. Then, the final position of the sugarcane nodes were determined according to the number of nodes to be identified and the sum of pixels of each 5 columns on both sides of the minimum points. The pixel column positions corresponding to the minima of the sum are the accurate nodes positions determined by the proposed algorithm. The experimental results show that the algorithm proposed in this paper has a single node identification rate of 100%, with an average time consumption of 0.15 s, and a position deviation of less than 0.34 mm; a double nodes identification rate of 98.5%, with an average time consumption of 0.21 s, and a position deviation of less than 0.42 mm. Compared with other nodes identification algorithms mentioned in this paper, it has higher identification rate and accuracy.

Tri-level handwritten text segmentation techniques for Gujarati language

Objectives: To improve the efficiency of tri-level segmentation tasks for handwritten Gujarati text. Methods: Using hybrid methods for tri-level segmentation, we have used line, word and character segmentation from the image. This study presents a segmentation paradigm that works with touching characters, slop of the line written on the page, character overlapping, etc. It evaluated on the dataset of 500+ images created by us on different writing sentences by different people. We have used the Horizontal projection technique for line segmentation, Scale-space technique for word segmentation and the Vertical projection technique for character segmentation. Findings: The experimental results show that the proposed method is more efficient for handwritten Gujarati text with diacritics. We have obtained the accuracy for character level segmentation is 82%, word-level is 90% and for the line-level segmentation is 87%. Novelty: We have designed a methodology to segment Gujarati handwritten text with diacritics at all three levels including characters, words and lines. Applications: We have proposed tri-level segmentation which is pre-processing task that can be used in any character recognition systems i.e. OCR. Keywords: Deep learning; trilevel segmentation; handwritten Gujarati text

Indonesian license plate recognition with improved horizontal-vertical edge projection

<p>License plate recognition (LPR) is one of the classical problems in the field of object recognition. Its application is very crucial in the automation of transportation system since it helps to recognise a vehicle identity, which information is stored in the license plate. LPR usually consists of three major phases: pre-processing, license plate localisation, optical character recognition (OCR). Despite being classical, its implementation faced with much more complicated problems in the real scenario. This paper proposed an improved LPR algorithm based on modified horizontal-vertical edge Projection. The method uses for detecting and localising the region of interest. It is done using the horizontal and vertical projection of the image. Related works proved that the modified horizontal-vertical edge projection is the simplest method to be implemented, yet very effective against Indonesian license plate. However, its performance gets reduced when specular reflection occurs in the sample image. Therefore, morphological operations are utilised in the pre-processing phase to reduce such effects while preserving the needed information. Eighty sample images which captured using various camera configurations were used in this research. Based on the experimental results, our proposed algorithm shows an improvement compared with the previous study and successfully detect 71 license plates in 80 image samples which results in 88.75% accuracy.</p>

Content-based medical image retrieval using a novel hybrid scattering coefficients - bag of visual words - DWT relevance fusion

Image content analysis plays a major role in image classification, retrieval, and indexing together with object and scene recognition. Numerous image content descriptors are proposed in the literature, but their high computational costs and lower-performance scores make them inappropriate for content-based medical image retrieval (CBMIR) for large medical image datasets. To overcome these drawbacks, a novel hybrid Scattering Coefficients - Bag of Visual Words – Discrete Wavelet Transform (SC-BoVW- DWT) relevance fusion algorithm is proposed for effective CBMIR. For preprocessing, resizing and contrast limited adaptive histogram equalization (CLAHE) are carried out. Then scattering transform (ST), BoVW and DWT are applied to extract texture features, visual features, and low-level features of the preprocessed images respectively. A hybrid Grey Wolf Optimization - Particle Swarm Optimization (GWO-PSO) approach is used for optimal feature selection. Finally, image fusion (IF) is carried out with a relevance fusion based Euclidean distance technique. Experiments based on three standard medical computer tomography image databases namely, EXACT-09, TCIA-CT and NEMA-CT, are carried out. The proposed hybrid method outperforms the existing techniques in terms of precision, F-score, and recall values. Improvement in the average rate of precision (ARP), average rate of recall (ARR) and F-score values of 6.02%, 2.82% and 3.57% respectively, for the EXACT-09 dataset and 3.55%, 1.84%, and 2.12% respectively for the TCIA-CT dataset are observed compared with the existing Scattering Transform- canonical correlation analysis vertical projection (ST-CCAv). For NEMA-CT an improvement of 0.21% (ARP) is obtained compared with the existing Histogram of compressed scattering coefficients (HCSC).

Improving method for deterministic treatment of double cross-slip in FCC metals under low homologous temperatures

Despite its importance in crystal plasticity, cross-slip has not yet been completely understood. The widely accepted approach understands cross-slip as a thermally-activated stochastic process, especially at higher homologous temperatures. An alternative approach is based on an experimentally supported observation that, especially at lower homologous temperatures, cross-slip becomes a deterministic stress-driven process governed by the line tension, by the applied stress, and by the short-range interactions among dislocations. The current challenge remains in understanding how the tip of the screw dislocation segment of changes the primary slip plane for the secondary slip plane. We suggest an explanation of this problem available by the mathematical theory of moving parametrized curves. For the dynamical change between the primary plane and the cross-slip plane, two methods are introduced and compared. They are referred to as the tilting projection method (denoted as TPM) and the vertical projection method (denoted as VPM). Main difference between them is how they regularize the motion of the dislocation across the edge between the primary and the cross-slip plane. After the motion strategy is set, this article explains that dislocation motion laws can be treated numerically by means of the flowing finite-volume method, including the redistribution of the discretization points ensuring a long-term stability of the algorithm. Particular computational results of cross-slip obtained by both approaches are mutually compared and evaluated. For the considered material setup of the FCC copper and nickel crystals, good agreement has been observed.