Search results

This paper aims to propose a system for defect detection in constructed elements that is able to indicate deformity positions. It also evaluates the defects in finishing materials of constructed building elements to support the subjective visual quality investigation of the aesthetics of an architectural work.

This strategy depends on defect features analysis that evaluates the defect value in digital images using digital image processing methods. The research uses the three-dimensional (3D) modeling techniques and image processing algorithms to generate a system that is able to perform some of the monitoring activities by computers. Based on the collected site scans, a 3D model is created for the building. Then, several images can be exported from the 3D model to investigate a specific element. Different image denoizing techniques are compared such as mean filter, median filter, Wiener filter and Split–Bregman iterations. The most efficient technique is implemented in the system. Then, the following six different methods are used for image segmentation to separate the concerned object from the background; color segmentation, region growing segmentation, histogram segmentation, local standard deviation segmentation, adaptive threshold segmentation and mean-shift cluster segmentation.

The proposed system is able to detect the cracks and defected areas in finishing works and calculate the percentage of the defected area compared to the total captured area in the photo with high accuracy.

The proposed system increases the precision of decision-making by decreasing the contribution of human subjective judgment. Investigation of different finishing surfaces is applied to validate the proposed system.

The purpose of this paper is to present a Differential Immune Clone Clustering Algorithm (DICCA) to solve image segmentation.

DICCA combines immune clone selection and differential evolution, and two populations are used in the evolutionary process. Clone reproduction and selection, differential mutation, crossover and selection are adopted to evolve two populations, which can increase population diversity and avoid local optimum. After extracting the texture features of an image and encoding them with real numbers, DICCA is used to partition these features, and the final segmentation result is obtained.

This approach is applied to segment all sorts of images into homogeneous regions, including artificial synthetic texture images, natural images and remote sensing images, and the experimental results show the effectiveness of the proposed algorithm.

The method presented in this paper represents a new approach to solving clustering problems. The novel method applies the idea two populations are used in the evolutionary process. The proposed clustering algorithm is shown to be effective in solving image segmentation.

The segmentation of printed circuit board (PCB) images is an important process in PCB inspection. The circuit traces, pads and vias in a PCB are dense and curved, and the PCB image obtained using different cameras or in different conditions may exhibit a large image gradient, which leads to inaccuracy and inefficiency in the PCB image segmentation. This paper aims to propose an improved local binary fitting level set method with prior graph cut, aiming to improve the accuracy and efficiency of the segmentation of PCB images obtained using different cameras or in different environments.

First, the paper constructs a 4-connected undirected graph using a given PCB image and classifies it based on the graph cut. Second, an adaptive initialization level set is implemented using the priori information obtained from the graph cut. Finally, the paper constructs a priori energy term using the prior information and introduces it into the energy function of the level set.

The approach results in an improved accuracy of segmentation in the context of a large gradient within the image. Experimental results demonstrate that the method can solve the deviation of artificially initialized level set from targets and improve the efficiency and accuracy of segmentation.

This study only considers level set method as the research object. Iteration of the level set method takes a long time for a given huge PCB picture, which makes it impossible to apply to scenes with high real-time requirements.

PCB image segmentation is an important process in the PCB inspection. Since template matching and morphology techniques are well-established, image segmentation quality has a significant impact on the accuracy of detection.

This paper studies the segmentation of PCB images, improves the efficiency and accuracy of segmentation and facilitates the subsequent applications, such as in the nondestructive testing of PCB.

The performance of three frequently used level set-based segmentation methods is examined for the purpose of defining features and boundary conditions for image-based Eulerian fluid and solid mechanics models. The focus of the evaluation is to identify an approach that produces the best geometric representation from a computational fluid/solid modeling point of view. In particular, extraction of geometries from a wide variety of imaging modalities and noise intensities, to supply to an immersed boundary approach, is targeted.

Two- and three-dimensional images, acquired from optical, X-ray CT, and ultrasound imaging modalities, are segmented with active contours, k-means, and adaptive clustering methods. Segmentation contours are converted to level sets and smoothed as necessary for use in fluid/solid simulations. Results produced by the three approaches are compared visually and with contrast ratio, signal-to-noise ratio, and contrast-to-noise ratio measures.

While the active contours method possesses built-in smoothing and regularization and produces continuous contours, the clustering methods (k-means and adaptive clustering) produce discrete (pixelated) contours that require smoothing using speckle-reducing anisotropic diffusion (SRAD). Thus, for images with high contrast and low to moderate noise, active contours are generally preferable. However, adaptive clustering is found to be far superior to the other two methods for images possessing high levels of noise and global intensity variations, due to its more sophisticated use of local pixel/voxel intensity statistics.

It is often difficult to know a priori which segmentation will perform best for a given image type, particularly when geometric modeling is the ultimate goal. This work offers insight to the algorithm selection process, as well as outlining a practical framework for generating useful geometric surfaces in an Eulerian setting.

The purpose of this paper is to propose a generic framework based on the colour and the texture features for colour‐textured image segmentation. The framework can be applied to any real‐world applications for appropriate interpretation.

The framework derives the contributions of colour and texture in image segmentation. Local binary pattern and an unsupervised k‐means clustering are used to cluster pixels in the chrominance plane. An unsupervised segmentation method is adopted. A quantitative estimation of colour and texture performance in segmentation is presented. The proposed method is tested using different mosaic and natural images and other image database used in computer vision. The framework is applied to three different applications namely, Irish script on screen images, skin cancer images and sediment profile imagery to demonstrate the robustness of the framework.

The inclusion of colour and texture as distributions of regions provided a good discrimination of the colour and the texture. The results indicate that the incorporation of colour information enhanced the texture analysis techniques and the methodology proved effective and efficient.

The novelty lies in the development of a generic framework using both colour and texture features for image segmentation and the different applications from various fields.

Though many segmentation methods have been published, few of them are developed especially for line scanned images. An ill‐illuminated line scanned (IILS) image tends to have a uniform intensity distribution in column direction while non‐uniform intensity distribution in the row direction. So, it is improper to segment IILS images using either a pixed threshold or threshold surface. In view of this, the purpose of this paper is to develop a segmentation method that is suitable for segmented IILS images.

To obtain satisfactory segmentation results, the illumination variation across the column of a line scanned image was taken into account and a column‐based segmentation method was developed. The method first calculates each column's standard deviation. Then a threshold value is automatically assigned to each column based on the derived values. Finally, by assembling each columns threshold value, a so‐called threshold line is formed. The method is threshold‐line segmentation method based on standard deviation (TLSTD).

The developed threshold‐line‐based segmentation method is compared with Otsu's fixed threshold segmentation method and Niblack's threshold‐surface‐based segmentation method. The results show that the threshold‐line‐based segmentation method is more suitable for segmenting IILS images.

Despite TLSTD outperforming Otsu's and Nilblack's segmentation methods, there are some limitations to it. The most obvious one is that the predetermined allowable deviation has influences on the integrality of the extracted flaws. Besides, since the proposed method is designed specifically for segmenting images captured by line scan cameras with a slant line light source, it is suitable for segmenting the kind of images only. In other words, the method shows no advantages in segment area scanned images.

Generally, the approach is useful in automated visual inspection where line scan cameras are employed.

The merit of the proposed method is that the slant of the line light source is now allowed. In other words, even if a grabbed line scanned image is unevenly illuminated, the proposed segmentation method is still able to successfully detect desired flaws.

Extraction of leaf region from the plant leaf images is a prerequisite process for species recognition, disease detection and classification and so on, which are required for crop management. Several approaches were developed to implement the process of leaf region segmentation from the background. However, most of the methods were applied to the images taken under laboratory setups or plain background, but the application of leaf segmentation methods is vital to be used on real-time cultivation field images that contain complex backgrounds. So far, the efficient method that automatically segments leaf region from the complex background exclusively for black gram plant leaf images has not been developed.

Extracting leaf regions from the complex background is cumbersome, and the proposed PLRSNet (Plant Leaf Region Segmentation Net) is one of the solutions to this problem. In this paper, a customized deep network is designed and applied to extract leaf regions from the images taken from cultivation fields.

The proposed PLRSNet compared with the state-of-the-art methods and the experimental results evident that proposed PLRSNet yields 96.9% of Similarity Index/Dice, 94.2% of Jaccard/IoU, 98.55% of Correct Detection Ratio, Total Segmentation Error of 0.059 and Average Surface Distance of 3.037, representing a significant improvement over existing methods particularly taking into account of cultivation field images.

In this work, a customized deep learning network is designed for segmenting plant leaf region under complex background and named it as a PLRSNet.

Brain tumor segmentation and classification is the interesting area for differentiating the tumorous and the non-tumorous cells in the brain and classifies the tumorous cells for identifying its level. The methods developed so far lack the automatic classification, consuming considerable time for the classification. In this work, a novel brain tumor classification approach, namely, harmony cuckoo search-based deep belief network (HCS-DBN) has been proposed. Here, the images present in the database are segmented based on the newly developed hybrid active contour (HAC) segmentation model, which is the integration of the Bayesian fuzzy clustering (BFC) and the active contour model. The proposed HCS-DBN algorithm is trained with the features obtained from the segmented images. Finally, the classifier provides the information about the tumor class in each slice available in the database. Experimentation of the proposed HAC and the HCS-DBN algorithm is done using the MRI image available in the BRATS database, and results are observed. The simulation results prove that the proposed HAC and the HCS-DBN algorithm have an overall better performance with the values of 0.945, 0.9695 and 0.99348 for accuracy, sensitivity and specificity, respectively.

The proposed HAC segmentation approach integrates the properties of the AC model and BFC. Initially, the brain image with different modalities is subjected to segmentation with the BFC and AC models. Then, the Laplacian correction is applied to fuse the segmented outputs from each model. Finally, the proposed HAC segmentation provides the error-free segments of the brain tumor regions prevailing in the MRI image. The next step is to extract the useful features, based on scattering transform, wavelet transform and local Gabor binary pattern, from the segmented brain image. Finally, the extracted features from each segment are provided to the DBN for the training, and the HCS algorithm chooses the optimal weights for DBN training.

The experimentation of the proposed HAC with the HCS-DBN algorithm is analyzed with the standard BRATS database, and its performance is evaluated based on metrics such as accuracy, sensitivity and specificity. The simulation results of the proposed HAC with the HCS-DBN algorithm are compared against existing works such as k-NN, NN, multi-SVM and multi-SVNN. The results achieved by the proposed HAC with the HCS-DBN algorithm are eventually higher than the existing works with the values of 0.945, 0.9695 and 0.99348 for accuracy, sensitivity and specificity, respectively.

This work presents the brain tumor segmentation and the classification scheme by introducing the HAC-based segmentation model. The proposed HAC model combines the BFC and the active contour model through a fusion process, using the Laplacian correction probability for segmenting the slices in the database.

Multi-view fuzzy clustering algorithms are not widely used in image segmentation, and many of these algorithms are lacking in robustness. The purpose of this paper is to investigate a new algorithm that can segment the image better and retain as much detailed information about the image as possible when segmenting noisy images.

The authors present a novel multi-view fuzzy c-means (FCM) clustering algorithm that includes an automatic view-weight learning mechanism. Firstly, this algorithm introduces a view-weight factor that can automatically adjust the weight of different views, thereby allowing each view to obtain the best possible weight. Secondly, the algorithm incorporates a weighted fuzzy factor, which serves to obtain local spatial information and local grayscale information to preserve image details as much as possible. Finally, in order to weaken the effects of noise and outliers in image segmentation, this algorithm employs the kernel distance measure instead of the Euclidean distance.

The authors added different kinds of noise to images and conducted a large number of experimental tests. The results show that the proposed algorithm performs better and is more accurate than previous multi-view fuzzy clustering algorithms in solving the problem of noisy image segmentation.

Most of the existing multi-view clustering algorithms are for multi-view datasets, and the multi-view fuzzy clustering algorithms are unable to eliminate noise points and outliers when dealing with noisy images. The algorithm proposed in this paper has stronger noise immunity and can better preserve the details of the original image.

Brain tumor is one of the most dangerous and life-threatening disease. In order to decide the type of tumor, devising a treatment plan and estimating the overall survival time of the patient, accurate segmentation of tumor region from images is extremely important. The process of manual segmentation is very time-consuming and prone to errors; therefore, this paper aims to provide a deep learning based method, that automatically segment the tumor region from MR images.

In this paper, the authors propose a deep neural network for automatic brain tumor (Glioma) segmentation. Intensity normalization and data augmentation have been incorporated as pre-processing steps for the images. The proposed model is trained on multichannel magnetic resonance imaging (MRI) images. The model outputs high-resolution segmentations of brain tumor regions in the input images.

The proposed model is evaluated on benchmark BRATS 2013 dataset. To evaluate the performance, the authors have used Dice score, sensitivity and positive predictive value (PPV). The superior performance of the proposed model is validated by training very popular UNet model in the similar conditions. The results indicate that proposed model has obtained promising results and is effective for segmentation of Glioma regions in MRI at a clinical level.

The model can be used by doctors to identify the exact location of the tumorous region.

The proposed model is an improvement to the UNet model. The model has fewer layers and a smaller number of parameters in comparison to the UNet model. This helps the network to train over databases with fewer images and gives superior results. Moreover, the information of bottleneck feature learned by the network has been fused with skip connection path to enrich the feature map.