Super Pixel Research Articles

Application research of denoising and super pixel algorithm in image processing

with the popularization and development of science and technology, mobile phone, tablet and computer has become the necessities of people, whether work or life, the emergence of science and technology, development and rich brought a whole new world for human civilization, including electronic information in time and space communication provides convenient conditions for people, especially the image processing technology, in the life is very broad. At present, smart phones have become extremely common, and users have a huge demand for images. Every link is inseparable from the formation, acquisition, transmission and acceptance of images. However, in every link, images will be more or less polluted by noise, resulting in users’ inability to obtain the desired image effect. However, if the noise is directly optimized or removed, the accuracy of the image will be affected. Therefore, the advanced noise removal technology plays a crucial role in the efficient use of the image. Image superpixel is to gather pixels with similar attributes into a region to represent the image instead of pixels, so as to reduce the order of magnitude of the image atomic structure and further reduce the complexity of the subsequent image processing algorithm, which provides the possibility for the real-time performance of the image processing algorithm.

Insulator detection and recognition of explosion based on convolutional neural networks

Unmanned aerial vehicles (UAVs) equipped with high definition (HD) cameras can obtain a large number of detailed inspection images. The insulator is an indispensable component in the transmission lines. Detecting insulator in image video quickly and accurately can provide a reliable basis for the ranging and the obstacle avoidance flight of UAV close to the tower and transmission line. At the same time, the insulator is a serious threat to the safety of the power grid due to the multiple faults of the insulator, and the computer technology should be fully utilized to diagnose the fault. Detection of the insulator images with the complex aerial background is implemented by constructing a convolutional neural network (CNN), which has the classic architecture of five modules of convolution and pooling, two modules of fully connected layers. In this paper, we propose a recognition algorithm for explosion fault based on saliency detection, which uses the trained network model to extract the features. Then, we put the saliency maps into a self-organizing feature map (SOM) network and build the mathematical module via super pixel segmentation, contour detection and other image processing methods. The test shows that the algorithm can reduce the error that may be caused by manual analysis. It also demonstrates that the detection of the insulator and the recognition of explosion fault can effectively improve the efficiency and intelligence level.

A Hierarchical Multiscale Super-Pixel-Based Classification Method for Extracting Urban Impervious Surface Using Deep Residual Network From WorldView-2 and LiDAR Data

High-resolution optical imagery can provide detailed information of urban land objects for impervious surface extraction, while airborne light detection and ranging (LiDAR) data can provide height features of land objects. Therefore, synergistic use of high-resolution imagery and LiDAR data is considered as an effective method to improve impervious surfaces extraction. In this paper, a novel hierarchical multiscale super-pixel-based classification method is proposed and applied to the urban impervious surfaces extraction from WorldView-2 and normalized digital surface model (nDSM) images derived from airborne LiDAR data. Three subsets in rural, rural–urban, and urban subsets are selected as the study areas. First, we split nonground and ground objects based on nDSM thresholds. Second, a hierarchical multiresolution segmentation method is used to generate nonground and ground super pixels. Then, we determine the multiscale input images based on the size of super pixels. Third, we construct optimal deep residual network (ResNet) and Spatial Pyramid Pooling (SPP-net) to train the model using multiscale input images. Finally, we use our deep models to predict hierarchically total super pixels in three subsets and generate the classification and impervious surfaces results. Our proposed method adopts hierarchical classification based on LiDAR nDSM height, which significantly improves the impervious surfaces extraction accuracies. Then, the deep residual network is applied further on multispectral and height fused data to extract urban impervious surfaces. Moreover, we propose an adaptive method to determine multiscale input images based on the segmentation of super pixels, which are inputs into the ResNet+SPP-net to train the deep model. Our proposed method reduces the uncertainty of multiscale input images and extracts better multiscale features. The results of the experiment show that our proposed method has a significant superiority to traditional pixel-based method and single scale method for urban impervious surfaces extraction.

Segmentation Algorithm of Medical Exercise Rehabilitation Image Based on HFCNN and IoT

Deep learning has achieved great success in the field of computer vision, and the precision in image classification and image detection has surpassed humans. Therefore, this paper combines deep learning and medical image segmentation, focusing on how to improve the accuracy and speed of segmentation algorithm of medical exercise rehabilitation image. Aiming at the shortcomings of traditional medical image recognition methods, a medical exercise rehabilitation image segmentation algorithm based on hierarchical features of convolutional neural networks is proposed, this paper calls it as hierarchical features of convolutional neural networks (HFCNN). The algorithm firstly samples the convolution output of multiple layers in the convolutional neural network to a unified scale and combines them to construct a hierarchical feature. This hierarchical feature combines the structural information of objects contained in the shallow layer of the network with the semantic information of objects contained in the deep layers of the network, so it has a strong ability to express. Secondly, the image can be segmented into multiple super pixels by the super pixel segmentation algorithm. The classifier is trained using the hierarchical features of the super pixel, and then the classification result is mapped back to the pixel. Finally, a fully connected conditional random field algorithm including one-potential potential energy and paired potential energy is constructed. The corresponding energy function is used to smooth the classification result of the pixel, and the regional consistency and continuity of the pixel mark are improved. Compared with many classical convolutional neural network algorithms, this algorithm not only accelerates the network convergence speed, shortens the training time, but also significantly improves the accuracy of segmentation algorithm of medical exercise rehabilitation image, showing good practical value.

Color image splicing localization algorithm by quaternion fully convolutional networks and superpixel-enhanced pairwise conditional random field.

Recently, fully convolutional network (FCN) has been successfully used to locate spliced regions in synthesized images. However, all the existing FCN-based algorithms use real-valued FCN to process each channel separately. As a consequence, they fail to capture the inherent correlation between color channels and the integrity of three channels. So, in this paper, quaternion fully convolutional network (QFCN) is proposed to generalize FCN to quaternion domain by replacing real-valued conventional blocks in FCN with quaternion conventional blocks. In addition, a new color image splicing localization algorithm is proposed by combining QFCNs and superpixel (SP)-enhanced pairwise conditional random field (CRF). QFCNs consider three different versions (QFCN32, QFCN16, and QFCN8) with different up-sampling layers. The SP-enhanced pairwise CRF is used to refine the results of QFCNs. Experimental results on three publicly available datasets demonstrate that the proposed algorithm outperforms the existing algorithms including some conventional algorithms and some deep learning-based algorithms.

Fast Recognition and Location of Target Fruit Based on Depth Information

In order to realize the recognition and localization of the target fruit more accurately and efficiently, an optimized graph-based recognition algorithm based on image depth information is proposed. Firstly, the gradient information of the depth image is obtained from the acquired depth apple image, the gradient vector is reflected from the 3D space to the 2D space, and then rotated 90 degrees clockwise, and the regular region forms the center of the vortex, which is the target fruit center to determine target fruit location. In the RGB space, the graph-based segmentation algorithm is introduced. By constructing the adaptive threshold, the mapping relationship between the number of centers and the parameters of k and min is established to find the target fruit region. Finally, scan the maximum radius of the super pixel area where the center of the circle is located, and fit the target fruit to realize the recognition and localization of the apple. The experimental results show that the target recognition rate is 96.61%, and the recognition time has been reduced significantly.

The Outlining of Agricultural Plots Based on Spatiotemporal Consensus Segmentation

The outlining of agricultural land is an important task for obtaining primary information used to create agricultural policies, estimate subsidies and agricultural insurance, and update agricultural geographical databases, among others. Most of the automatic and semi-automatic methods used for outlining agricultural plots using remotely sensed imagery are based on image segmentation. However, these approaches are usually sensitive to intra-plot variability and depend on the selection of the correct parameters, resulting in a poor performance due to the variability in the shape, size, and texture of the agricultural landscapes. In this work, a new methodology based on consensus image segmentation for outlining agricultural plots is presented. The proposed methodology combines segmentation at different scales—carried out using a superpixel (SP) method—and different dates from the same growing season to obtain a single segmentation of the agricultural plots. A visual and numerical comparison of the results provided by the proposed methodology with field-based data (ground truth) shows that the use of segmentation consensus is promising for outlining agricultural plots in a semi-supervised manner.

Saliency object detection: integrating reconstruction and prior

To remedy some challenging cases in saliency detection such as complex background and multiple objects. A new saliency object detection approach is proposed via integrating reconstruction and prior knowledge. This paper first segments each image into super pixels using over-segmentation algorithm. Then, the reconstruction saliency map and prior saliency map are generated by reconstruction and prior, respectively. The reconstruction involves dense reconstruction and sparse reconstruction. When the saliency object appears on the image boundaries, the detection can be more accurate via dense reconstruction. In addition, if there is complex background in natural scene image, the sparse reconstruction can be more robust and suppress the background effectively. The prior adopts background prior and center prior, which can highlight the saliency object uniformly. The reconstruction saliency map and prior saliency map are nonlinearly integrated to generate the final saliency map. The proposed method is compared with the other five state-of-the-art algorithms based on comprehensive metrics. The experimental results demonstrate that the proposed algorithm has superior saliency detection performance and low average elapsing time.

Pixel Connectivity Approach for Colour Reconstruction of 3D Brain Image

Today picture getting ready and uses in perspective of RGB-D data have colossal excitement for various research districts. As to, shading alteration, significance information, broad depiction of 3-Estimation scenes in RGB Significance data gives supportive experiences from another estimation to detach substance in close tones. Here, a novel technique is discussed to decide such mixed, lost, and tumult level addition of pixel issues of the RGB-D structure and has been replicated using MATLAB programming. The upside of MATLAB is, it is for the most part available, continually invigorated and has more broad reach. In our methodology, mixed pixel zones are distinguished using super pixel division of shading and significance, and harden them to lost pixel locale. Likewise, a hybrid channel is used to clear confusion level of pixels. Exploratory results exhibits the proposed system gives better execution on PSNR, SNR, Gain.

High extinction ratio super pixel for long wavelength infrared polarization imaging detection based on plasmonic microcavity quantum well infrared photodetectors

Polarization imaging detection has its unique advantage in discriminating the man-made objects from natural objects. Grating integrated super pixel for polarization imaging detection can simultaneously obtain the first three elements of the Stokes vector, which is the trend of infrared polarization imaging detection in recent years. Here, we demonstrate the first super pixel for long wavelength infrared polarization imaging detection with the extinction ratio of its four polarization directions more than 100. The measured highest polarization extinction ratio is as high as 136, which is the highest reported value of long wavelength infrared polarization imaging detection super pixel. The mechanism is attributed to the excellent mode selectivity of plasmonic microcavity according to the results of three-dimensional theoretical simulation. The experimental responses of the super pixel with four polarization directions are in good agreement with the Malus’ Law. In addition, the super pixel can accurately resolve the Stokes parameters at the same time. It is expected to develop the super pixel into a new generation of practical high-polarization-discriminating long wavelength infrared focal plane array.

A Method based on Super Pixel and Artificial Neural Network for Lung Detection from CT images

Tıbbi görüntülerden doku veya organların otomatik olarak tespit edilmesi bilgisayarlı görü alanının önemli çalışma konularından birisidir. Bu çalışmada bilgisayarlı tomografi (BT) görüntülerinden akciğerin otomatik olarak tespiti için bir yöntem önerilmiştir. Önerilen yöntem süper pikselleri kullanan yapay sinir ağları (YSA) üzerinde temellendirilmiştir ve klinik karar destek sistemleri için ilk aşama olarak kullanılması hedeflenmektedir. Yöntemin başarım incelemesi National Lung Screening Trial (NLST) veri tabanındaki BT görüntüleri üzerinde gerçekleştirilmiştir.

Cross-View Multi-Lateral Filter for Compressed Multi-View Depth Video.

Multi-view depth is crucial for describing positioning information in 3D space for virtual reality, free viewpoint video, and other interaction- and remote-oriented applications. However, in cases of lossy compression for bandwidth limited remote applications, the quality of multi-view depth video suffers from quantization errors, leading to the generation of obvious artifacts in consequent virtual view rendering during interactions. Considerable efforts must be made to properly address these artifacts. In this paper, we propose a cross-view multi-lateral filtering scheme to improve the quality of compressed depth maps/videos within the framework of asymmetric multi-view video with depth compression. Through this scheme, a distorted depth map is enhanced via non-local candidates selected from current and neighboring viewpoints of different time-slots. Specifically, these candidates are clustered into a macro super pixel denoting the physical and semantic cross-relationships of the cross-view, spatial and temporal priors. The experimental results show that gains from static depth maps and dynamic depth videos can be obtained from PSNR and SSIM metrics, respectively. In subjective evaluations, even object contours are recovered from a compressed depth video. We also verify our method via several practical applications. For these verifications, artifacts on object contours are properly managed for the development of interactive video and discontinuous object surfaces are restored for 3D modeling. Our results suggest that the proposed filter outperforms state-of-the-art filters and is suitable for use in multi-view color plus depth-based interaction- and remote-oriented applications.

Superpixel-Level CFAR Detectors for Ship Detection in SAR Imagery

Synthetic aperture radar (SAR) is one of the most widely employed remote sensing modalities for large-scale monitoring of maritime activity. Ship detection in SAR images is a challenging task due to inherent speckle, discernible sea clutter, and the little exploitable shape information the targets present. Constant false alarm rate (CFAR) detectors, utilizing various sea clutter statistical models and thresholding schemes, are near ubiquitous in the literature. Very few of the proposed CFAR variants deviate from the classical CFAR topology; this letter proposes a modified topology, utilizing superpixels (SPs) in lieu of rectangular sliding windows to define CFAR guardbands and background. The aim is to achieve better target exclusion from the background band and reduced false detections. The performance of this modified SP-CFAR algorithm is demonstrated on TerraSAR-X and SENTINEL-1 images, achieving superior results in comparison to classical CFAR for various background distributions.

An Integrated Technique for Image Forgery Detection using Block and Keypoint Based Feature Techniques

The invention of the net has introduced the unthinkable growth and developments within the illustrious analysis fields like drugs, satellite imaging, image process, security, biometrics, and genetic science. The algorithms enforced within the twenty first century has created the human life more leisurely and secure, however the protection to the first documents belongs to the genuine person is remained as involved within the digital image process domain. a replacement study is planned during this analysis paper to discover. The key plan in the deliberate take a look at and therefore the detection of the suspected regions are detected via the adaptive non-overlapping and abnormal blocks and this method is allotted exploitation the adaptive over-segmentation algorithmic rule. The extraction of the feature points is performed by playacting the matching between every block and its options. The feature points are step by step replaced by exploitation the super pixels within the planned Forgery Region Extraction algorithm then merge the neighboring obstructs that have comparative local shading decisions into the element squares to encourage the brought together districts; at last, it applies the morphological activity to the bound together areas to ask the recognized falsification districts The planned forgery detection algorithmic rule achieves far better detection results even below numerous difficult conditions the sooner strategies all told aspects. We have analyzed the results obtained by the each SIFT and SURF and it is well-tried that the planned technique SURF is giving more satisfactory results by both subjective and objective analysis.

Use of Spectral Clustering Combined with Normalized Cuts (N-Cuts) in an Iterative k-Means Clustering Framework (NKSC) for Superpixel Segmentation with Contour Adherence

Superpixel segmentation methods are generally used as a pre-processing step to speed up image processing tasks. They group the pixels of an image into homogeneous regions while trying to respect existing contours. In this paper, we propose a fast Superpixels segmentation algorithm with Contour Adherence using spectral clustering, combined with normalized cuts in an iterative k-means clustering framework. It produces compact and uniform superpixels with low computational costs. Normalized cut is adapted to measure the color similarity and space proximity between image pixels. We have used a kernel function to estimate the similarity metric. Kernel function maps the pixel values and coordinates into a high dimensional feature space. The objective functions of weighted K-means and normalized cuts share the same optimum point in this feature space. So it is possible to optimize the cost function of normalized cuts by iteratively applying simple K-means clustering algorithm. The proposed framework produces regular and compact superpixels that adhere to the image contours. On segmentation comparison benchmarks it proves to be equally well or better than the state-of-the-art super pixel segmentation algorithms in terms of several commonly used evaluation metrics in image segmentation. In addition, our method is computationally very efficient and its computational complexity is linear.

SUPER PIXEL USING RANDOM WALK SEGMENTATION

Objectives: An innovative image superpixel segmentation approach run on the test image to acquire the probabilities of every pixel and performs grouping in light of shading similarities and spatial nearness of the pixels in histopathologic images. LRW algorithm using self-circles has the benefits of portioning the weak limits and confounded surface areas extremely fit. Method of analysis: This technique starts with instating the seed locations and tracks the LRW algorithm on the test image to acquire the probabilities of every pixel. At that point, the limits of starting superpixels is acquired by the probabilities what’s more, the drive time. Findings: At the point when the outcomes are assessed, it has been watched that the superpixel strategy has a positive commitment to both the segmentation achievement and the running time. The execution of superpixel is enhanced using moving the focus places of super pixels then separating the huge superpixels into little with the proposed improvement procedure. The exploratory outcomes have shown that our technique accomplishes preferable execution over past superpixel approaches. Application/Improvements: To detect the distance objects as a test image, and through this method, we can analyze and provide an optimal solution where time is constraint. Keywords: Features Classification Forest, Grey Level Co-Occurrence Matrix, Single Value Decomposition

Saliency Region Detection Method Based on Background and Spatial Position

Saliency region detection methods have become one of the hotspots in the field of image processing as an important method to improve the real-time and accurate analysis of massive data. Integrating more effective prior knowledge is a viable direction for improving the performance of saliency region detection methods. Most of the methods based on background prior and boundary connectivity prior assume the boundary area of the image as the background, by restraining the background to highlight the salient area. When the boundary area of the image does not describe the background well (such as a large difference in border area features), if the entire frame of the image is put together to compute the background feature, the calculation of the background feature will be inaccurate. In view of the above shortcomings, this paper proposed a saliency region detection method based on background and spatial position. This method carried on the image boundary super pixel clustering, determined the background feature according to the clustering center, and used the difference between the super pixel on the image and the background super pixel, and its spatial position to calculate the salient of the super pixels. This approach used MATLAB to program and experiment. The method was compared with a series of the state-of-the-art methods. The AUC of proposed algorithm reaches 0.839, and the MAE is 0.220, showing the effectiveness of the proposed algorithm.

Semantic segmentation based on fusion of features and classifiers

This paper proposes a feed forward architecture algorithm using fusion of features and classifiers for semantic segmentation. The algorithm consists of three phases: Firstly, the features from hierarchical convolutional neural network (CNN) and the features based on region are extracted and fused on super pixel level; secondly, multiple classifiers of Softmax, XGBoost and Random Forest are ensemble to compute the per-pixel class probabilities; at last, a fully connected conditional random field is employed to enhance the final performance. The hierarchical features contain more global evidence and the region features contain more local evidence. So the fusion of these two features is expected to enhance the feature representation ability. In classification phase, integrating multiple classifiers aims to improve the generalization ability of classification algorithms. Experiments are conducted on Sift-Flow datasets by our proposed methods with competitive labeling accuracy.

TS2uRF: A New Method for Sharpening Thermal Infrared Satellite Imagery

Thermal infrared (TIR) imagery is normally acquired at coarser pixel resolution than that of shortwave sensors on the same satellite platform. TIR resolution is often not suitable for monitoring crop conditions of fragmented farming lands, e.g., the accurate estimates of evapotranspiration (ET) based on surface energy balance from remote sensing for irrigation water management. Consequently, thermal sharpening techniques have been developed to sharpen TIR imagery to a shortwave band pixel resolution. However, most methods concentrate on the visual effects of the thermal sharpened images, and they treat the pixels as independent samples without considering their spatial context, which can give rise to adverse effects such as artifacts. In this work, a new thermal sharpening method called TS2uRF is proposed. The potential of superpixels (SP) combined with regression random forest (RRF) have been used to augment the spatial resolution of the Landsat 8 TIR (100 m) imagery to their visible (VIS) spatial resolution (30 m). The SP has allowed the contextual information on the land cover to be integrated, and RRF has allowed the relationship between five spectral indices and TIR data to be integrated into a single model. The TIR sharpened images obtained using the TS2uRF were compared with images obtained using the TsHARP, one of the most classic thermal sharpening techniques, evaluating the root-mean-square error (RMSE) and structural similarity index (SSIM) for measuring image quality. In all of the cases evaluated, the RMSE and SSIM of the images sharpened using the TS2uRF method outperform those obtained using TsHARP. In particular, the TS2uRF method has an average error of 1.14 °C (RMSE) lower than TsHARP, regarding SSIM, TS2uRF outperforms TsHARP on average by 0.218 . From the visual comparison, it has been shown that the TS2uRF methodology avoids the artifacts that appear in the enhanced images using the TsHARP method.

Fast Stereo Matching Fully Utilizing Super-Pixels

In this paper, we propose a depth image generation method by stereo matching on super-pixel (SP) basis. In the proposed method, block matching is performed only at the center of the SP, and the obtained disparity is applied to all pixels of the SP. Next, in order to improve the disparity, a new SP-based cost filter is introduced. This filter multiplies the matching cost of the surrounding SP by a weight based on reliability and similarity and sums the weighted costs of neighbors. In addition, we propose two new error checking methods. One-way check uses only a unidirectional disparity estimation with a small amount of calculation to detect errors. Cross recovery uses cross checking and error recovery to repair lacks of objects that are problematic with SP-based matching. As a result of the experiment, the execution time of the proposed method using the one-way check was about 1/100 of the full search, and the accuracy was almost equivalent. The accuracy using cross recovery exceeded the full search, and the execution time was about 1/60. Speeding up while maintaining accuracy increases the application range of depth images.