Bi-level Thresholding Research Articles

A Novel Percentage Split Distribution Method for Image Thresholding

Traditional image thresholding methods are found to be effective for bi-level thresholding, but computationally expensive for multi-level thresholding. Most of these methods do not precisely segment images with zero root mean square error. Nevertheless these methods fail to select quality thresholds with the increased number of segmentations. To overcome these issues, this paper proposes a novel and fastest image thresholding method ‘percentage split distribution’ to segment images into regions of relevant groups. The method effectively computes the globally optimized thresholds through percentage-wise segmentation of image pixels. To analyze the performance of the proposed method, nine bench mark grayscale images are taken and validated against SSIM, PSNR and time complexity measures. The experimental results show that the proposed method is superior to other thresholding based image segmentation methods in terms of high PSNR and SSIM values resulted with reduced time complexity. The experimental results also proclaim that the thresholds generated by the proposed method are effective and are of equal intervals. Moreover, the proposed method is found to be effective, simple, easy to implement and can be effectively used for real-time applications.

An efficient Harris hawks-inspired image segmentation method

Segmentation is a crucial phase in image processing because it simplifies the representation of an image and facilitates its analysis. The multilevel thresholding method is more efficient for segmenting digital mammograms compared to the classic bi-level thresholding since it uses a higher number of intensities to represent different regions in the image. In the literature, there are different techniques for multilevel segmentation; however, most of these approaches do not obtain good segmented images. In addition, they are computationally expensive. Recently, statistical criteria such as Otsu, Kapur, and cross-entropy have been utilized in combination with evolutionary and swarm-based strategies to investigate the optimal threshold values for multilevel segmentation. In this paper, an efficient methodology for multilevel segmentation is proposed using the Harris Hawks Optimization (HHO) algorithm and the minimum cross-entropy as a fitness function. To substantiate the results and effectiveness of the HHO-based method, it has been tested over a benchmark set of reference images, with the Berkeley segmentation database, and with medical images of digital mammography. The proposed HHO-based solver is verified based on other comparable optimizers and two machine learning algorithms K-means and the Fuzzy IterAg. The comparisons were performed based on three groups. This first one is to provide evidence of the optimization capabilities of the HHO using the Wilcoxon test, and the second is to verify segmented image quality using the PSNR, SSIM, and FSIM metrics. Then, the third way is to verify the segmented image comparing it with the ground-truth through the metrics PRI, GCE, and VoI. The experimental results, which are validated by statistical analysis, show that the introduced method produces efficient and reliable results in terms of quality, consistency, and accuracy in comparison with the other methods. This HHO-based method presents an improvement over other segmentation approaches that are currently used in the literature.

A fuzzy adaptive gravitational search algorithm for two-dimensional multilevel thresholding image segmentation

Two-dimensional (2D) multilevel thresholding is an important technique for noisy image segmentation which has drawn much attention during the past few years. The conventional image segmentation methods are efficient for 2D bi-level thresholding. However, the computational complexity grows exponentially when extended to 2D multilevel thresholding since they search the optimal thresholds by exhaustive strategy. To tackle this problem, a fuzzy adaptive gravitational search algorithm (FAGSA) using Tsallis entropy as its objective function has been presented to find the optimal 2D multilevel thresholds in this paper. In the FAGSA, fuzzy logic controllers are designed to tune the control parameters. The state-of-the-art heuristic algorithms are compared with this proposed algorithm. Both test images and noisy images are utilized in the experiments to evaluate the performance of the involved algorithms. The experimental results significantly demonstrate the superiority of our algorithm in terms of the objective function value, image quality measures and time consumption.

An efficient krill herd algorithm for color image multilevel thresholding segmentation problem

The conventional thresholding methods are very efficient for bi-level thresholding, but the computational complexity may be excessively high for color image multilevel thresholding. Color image multilevel thresholding segmentation can be considered as a constrained optimization problem, therefore swarm intelligence algorithms are widely used to reduce the complexity. In this paper, an efficient krill herd (EKH) algorithm is proposed to search optimal thresholding values at different level for color images and the Otsu’s method, Kapur’s entropy and Tsallis entropy are employed as objective functions. Seven different algorithms, KH without any genetic operators (KH I), KH with crossover operator (KH II), KH with crossover and mutation operators (KH IV), modified firefly algorithm (MFA), modified grasshopper optimization algorithm (MGOA), bat algorithm (BA) and water cycle algorithm (WCA), are compared with the EKH algorithm. Experiments are performed on ten color benchmark images in terms of optimal threshold values, objective values, PSNR, SSIM and standard deviation of the objective values at different levels. The experimental results show that the presented EKH algorithm is superior to the other algorithms for color image multilevel thresholding segmentation. On the other hand, Kapur’s entropy is found to be more accurate and robust for color image multilevel thresholding segmentation.

Modified thermal exchange optimization based multilevel thresholding for color image segmentation

This paper proposes a multi-threshold image segmentation method based on modified thermal exchange optimization (TEO). Although it is efficient and gives excellent result in the case of bi-level thresholding, but it takes a lot of computation when the number of threshold increases. To overcome this problem, the TEO algorithm is applied in this search area for searching the optimal thresholds. The Levy flight algorithm is employed to modify the original TEO and balance the exploration and exploitation of TEO. Experiments are conducted between six state-of-the-art metaheuristic algorithms and the proposed one. The benchmark functions, color nature images and stellite images are used in the experiment to test the performance of the algorithms involved. Qualitative experimental results show that the proposed segmentation approach has fewer iterations and higher segmentation accuracy.

Improving retinal vessel segmentation with joint local loss by matting

Besides the binary segmentation, many retinal image segmentation methods also produce a score map, where a nonnegative score is assigned for each pixel to indicate the likelihood of being a vessel. This observation inspires us to propose a new approach as a post-processing step to improve existing methods by formulating segmentation as a matting problem. A trimap is obtained via a bi-level thresholding of the score map from existing methods, which is instrumental in focusing the attention to pixels of these unknown areas. A dedicated end-to-end matting algorithm is further developed to retrieve those vessel pixels in the unknown areas, and to produce the final vessel segmentation by minimizing global pixel loss and local matting loss. Our approach is shown to be particularly effective in rescuing thin and tiny vessels that may lead to disconnections of vessel fragments. Moreover, it is observed that our approach is capable of improving the overall segmentation performance across a broad range of existing methods.

Masi entropy based multilevel thresholding for image segmentation

A new multilevel thresholding based image segmentation technique is developed which utilizes Masi entropy as an objective function. Thresholding is an important image segmentation technique. It may be divided into two types such as bi-level and multilevel thresholding. Bi-level thresholding uses a single threshold to classify an image into two classes: object and the background. For an image containing a single object in a distinct background, bi-level thresholding can be successfully used for segmentation. But in case of complex images containing multiple objects, bi-level thresholding often fails to give satisfactory segmentation. In such cases, multilevel thresholding is generally preferred over bi-level thresholding. However, computational complexity of multilevel thresholding increases very rapidly with increasing number of thresholds. Metaheuristic algorithms are generally used to optimize the threshold searching process to reduce the computational complexity involved in multilevel thresholding. In this paper, Particle Swarm Optimization (PSO) along with Masi entropy is proposed for multilevel thresholding based image segmentation. The proposed technique is evaluated using a set of standard test images. The proposed technique is compared with the recently proposed Dragonfly Algorithm (DA) based technique that uses Kapur’s entropy as objective function. The proposed technique is also compared with PSO based technique that uses minimum cross entropy (MCE) as objective function. The quality of the segmented images is measured using Mean Structural SIMilarity (MSSIM) index and Peak Signal-to-Noise Ratio (PSNR). The experimental results suggest that the proposed technique outperforms Kapur’s entropy and gives very competitive result when compared with the MCE based technique. Further, computational complexity of multilevel thresholding is also greatly reduced.

Image thresholding segmentation method based on minimum square rough entropy

Image thresholding based on rough entropy is an efficient image segmentation technique. The optimal thresholds of the existing exponential and logarithmic rough entropy thresholding segmentation algorithms have unclear physical meaning. In this paper, a new form of square rough entropy is defined to measure the roughness in an image, and the corresponding image thresholding segmentation algorithm is proposed. The novel square rough entropy has good properties and simple computation. In the proposed thresholding algorithm, the optimal threshold is at the boundary between the object and the background of an image. This process is consistent with the expectation of image bi-level thresholding segmentation. To effectively granulate the image, a granule size selection method based on the homogeneity histogram is proposed, which is helpful in taking care of small objects and local variations of the images. The proposed thresholding method is evaluated by comparison with other three existing rough entropy based thresholding methods and three state-of-the-art image thresholding methods both qualitatively and quantitatively, using the natural images, the non-destructive testing images and an infrared video sequence from the OTCBVS Benchmark Data set. The comparison confirms the effectiveness of the proposed algorithm that not only is simple in form and clear in meaning, but also has good segmentation effect.

Multi-level image thresholding based on social spider algorithm for global optimization

Thresholding is one of the simplest and popular technique for segmenting images. Maximum between-class variance (Otsu’s) method is one of the well-known and widely used method in case of segmentation. Not only Otsu could be used for bi-level thresholding but also it could be extended to multi-level image thresholding. Finding the optimum threshold values in multi-level case is very time consuming process, thus optimization algorithm can deal with this problem. In this paper social spider algorithm for global optimization has been used for maximizing the between-class variance to carry out multi-level image thresholding. Experimental outcomes have demonstrated that the proposed method is capable of estimating threshold values and yield satisfying outcome.

An adaptive differential evolution algorithm to optimal multi-level thresholding for MRI brain image segmentation

Segmentation is an important method for MRI medical image analysis as it can provide the radiologists with noninvasive information about a patient that is crucial to the diagnostic process. The efficiency of such a computer-aided diagnosis system relies on the accuracy of an adopted image segmentation method. Multi-level thresholding is a segmentation method that has been widely adopted in medical image analysis in recent studies, where selecting the optimal thresholds has a pivotal role in determining the efficiency and the accuracy of the segmentation algorithm. While some well-known methods, such as Kapur’s and Otsu’s, are proven effective for bi-level thresholding, multi-level thresholding remains a challenge as it is computationally expensive. Evolutionary algorithms, such as Differential Evolution (DE), have the potential to address this problem, as they can find sufficiently good solutions with manageable computational effort. While a number of DE solutions have been proposed for multi-level thresholding, they are not stable, in that, when the number of thresholds increases, the algorithm efficiency decreases due to the imbalance between exploration and exploitation. In this paper, we propose a DE solution that achieves a good balance between exploration and exploitation through a new adaptive approach and new mutation strategies. The new adaptive approach can generate optimal solutions in assigning populations by measuring the quality of candidate solutions to evaluate the efficiency of different parts of the proposed DE algorithm. The new mutation methods harness Mantegna Lévy and Cauchy distributions, as well as Cotes’ Spiral to improve global search, and to further balance between exploitation and exploration. We further experimentally compare the proposed DE algorithm, referred to as Adaptive Differential Evolution with Lévy Distribution (ALDE), against three DE benchmark algorithms on T2 weighted MRI brain images. Our results show that ALDE can, not only obtain optimal thresholds at a reasonable computational cost, but more importantly, clearly outperforms the benchmark algorithms.

Bi-Level Thresholding

In gesture recognition, one challenge that researchers and developers face is the need for recognition strategies that mediate between false positives and false negatives. In this article, we examine bi-level thresholding, a recognition strategy that uses two thresholds: a tighter threshold limits false positives and recognition errors, and a looser threshold prevents repeated errors (false negatives) by analyzing movements in sequence. We first describe early observations that led to the development of the bi-level thresholding algorithm. Next, using a Wizard-of-Oz recognizer, we hold recognition rates constant and adjust for fixed versus bi-level thresholding; we show that systems using bi-level thresholding result in significantly lower workload scores on the NASA-TLX and significantly lower accelerometer variance when performing gesture input. Finally, we examine the effect that bi-level thresholding has on a real-world dataset of wrist and finger gestures, showing an ability to significantly improve measures of precision and recall. Overall, these results argue for the viability of bi-level thresholding as an effective technique for balancing between false positives, recognition errors, and false negatives.

Modified Grasshopper Algorithm-Based Multilevel Thresholding for Color Image Segmentation

Multilevel thresholding is an important approach for image segmentation which has drawn much attention during the past few years. The Tsallis entropy method is implemented for its effectiveness and simplicity. Although it is efficient and gives an excellent result in the case of bi-level thresholding, its evaluation becomes complexity when the number of thresholds increases. To overcome the problem, the metaheuristic algorithms are applied in this search area for searching the optimal thresholds. In this paper, a modified grasshopper optimization algorithm (GOA) is adopted to render multilevel Tsallis cross entropy more practical and reduce the complexity. The Levy flight algorithm is employed to modify the original GOA and balance the exploration and exploitation of the GOA. Experiments are conducted between five state-of-the-art metaheuristic algorithms and the proposed one. In addition, the proposed approach is compared with thresholding techniques depending on between-class variance (Otsu) method and the Renyi entropy function. Both real life images and plant stomata images are used in the experiments to test the performance of the algorithms involved. Qualitative experimental results show that the proposed segmentation approach has a fewer iterations and a higher segmentation accuracy.

An Improved Bi-Level Thresholding Based Uncertainty Evaluation for Speech Enhancement in Non-Stationary Noises

This paper proposes a new speech enhancement framework to improve the quality of speeches recorded under adverse acoustic environments based on the speech presence uncertainty. Since the uncertainty evaluation gives a more and clear discrimination about the speech and noise, this paper proposes a new uncertainty evaluation mechanism as a preprocessing mechanism to the noise suppression methods. This mechanism relates with energies of a noisy speech signal and classifies the speech segments and noise segments more perfectly. In addition to the quality enhancement, this approach also reduces the unnecessary computational burden over the speech processing system. Extensive simulations are carried out over the speech signals with different types of non-stationary noises like babble noise, exhibition noise, restaurant noise and train station noises and the performance is measured with the performance metrics namely the Output SNR, AvgSegSNR, PESQ and COMP. The comparative analysis of proposed approach over the conventional approaches shows an outstanding performance in all environments.

A novel automatic dam crack detection algorithm based on local-global clustering

Dam crack detection is necessary to ensure the safety of dams. However, traditional detection methods always perform poorly, with a low detection rate and high false alarm rate, due to the complex underwater environment. In this paper, a novel automatic dam crack detection algorithm (CrackLG) is proposed based on local-global clustering analysis that can find cracks on dam surfaces accurately and quickly using images as well as reduce human subjectivity. First, an image shot of an underwater dam surface is divided into non-overlapping image blocks after pre-processing. Then, image blocks containing crack pixels are identified by local clustering analysis. Second, the image is binarized by adaptive bi-level thresholding based on the local gray intensity. Meanwhile, some noise is removed based on the computed optimal threshold. After extracting global 3-D features, final crack regions are obtained by global clustering analysis. The advantage of CrackLG is that the threshold for realizing image binarization is self-adaptive. Additionally, it can automatically perform crack detection without human supervision. The simulation and comparison show that the proposed CrackLG method is more effective for underwater dam crack detection.

Optimization of Charge/Discharge Coordination to Satisfy Network Requirements Using Heuristic Algorithms in Vehicle-to-Grid Concept

Image thresholding is the most crucial step in microscopic image analysis to distinguish bacilli objects causing of tuberculosis disease. Therefore, several bi-level thresholding algor ...

A Novel Approach for Bi-Level Segmentation of Tuberculosis Bacilli Based on Meta-Heuristic Algorithms

Image thresholding is the most crucial step in microscopic image analysis to distinguish bacilli objects causing of tuberculosis disease. Therefore, several bi-level thresholding algor ...

A novel automated absolute intensity difference based technique for optimal MR brain image thresholding

The traditional Otsu technique for thresholding based on 2D histogram does not give accurate results. A bi-level thresholding case partitions the 2D histogram into four quadrants. This converts the global probability distribution to a probability distribution of two regions only resulting in partial loss of information. The brain regions (white matter (WM), gray matter (GM), cerebrospinal fluid (CSF)) have closely defined gray values, which require the intensity difference information for thresholding. So we propose a novel automated absolute intensity difference based (AIDB) technique for optimal MR brain image thresholding using adaptive coral reef optimization (ACRO). The basic idea is to extract the intensity difference information of the brain image from the 2D histogram matrix. A first-hand objective function, depending on the classical between class variance concepts, is investigated, which is maximized by ACRO. The key achievements of our technique are (i) improved thresholding results, (ii) obtaining more homogenous regions of the image and (iii) obtaining more precise shape of the edges. The proposed technique is tested with one hundred slices of the axial T2 – weighted MR brain images of Harvard medical dataset. The Otsu method for multilevel thresholding using the 2D histogram is considered for a comparison. Several performance measures are considered for the evaluation. It is observed that our technique outperforms the other standard methods.

Multi-Level Thresholding for Image Segmentation With Swarm Optimization Algorithms

Image segmentation is an important problem for image processing. The image processing applications are generally affectedfromthe segmentation success. There is noany image segmentation method which gives good results for all sorts of images. That’s why there are many approaches and methods forimage segmentationin the literature. And one of the most used is the thresholding technique. Thresholding techniques can be categorized into two topics: bi-level and multi-level thresholding. Bi-level thresholding technique has one threshold value which separates the image into two groups. However, multi-level thresholding technique uses n threshold values where n greater than one. In this paper, two swarm optimization algorithms (Particle Swarm Optimization, PSO and Cat Swarm Optimization, CSO) are applied on finding the optimum threshold values for the multi-level thresholding. In literature, there are some minimization or maximization functions to find the best threshold values for thresholding problem. Some of these methods are: Tsalli’s Entropy, Kapur’s Entropy, Renyi’s Entropy, Otsu’s Method (within class variance/between class variance), the Minimum Cross Entropy Thresholding (MCET) etc.In this work, Otsu’s (within class variance) method, which is one of these popular functions,is used as the fitness function of algorithms.In the experiments, five real images are segmented by usingParticle Swarm Algorithm and Cat Swarm Optimization Algorithms. The performances of the swarm algorithms on multi-level thresholding problem arecompared with Peak Signal-to-Noise Ratio (PSNR) and fitness function (FS) values. As a result, the PSO yields better performance than CSO.

Image segmentation by minimum cross entropy using evolutionary methods

The segmentation of digital images is one of the most important steps in an image processing or computer vision system. It helps to classify the pixels in different regions according to their intensity level. Several segmentation techniques have been proposed, and some of them use complex operators. The techniques based on thresholding are the easiest to implement; the problem is to select correctly the best threshold that divides the pixels. An interesting method to choose the best thresholds is the minimum cross entropy (MCET), which provides excellent results for bi-level thresholding. Nevertheless, the extension of the segmentation problem into multiple thresholds increases significantly the computational effort required to find optimal threshold values. Each new threshold adds complexity to the formulation of the problem. Classic methods for image thresholding perform extensive searches, while new approaches take advantage of heuristics to reduce the search. Evolutionary algorithms use heuristics to optimize criteria over a finite number of iterations. The correct selection of an evolutionary algorithm to minimize the MCET directly impacts the performance of the method. Current approaches take a large number of iterations to converge and a high rate of MCET function evaluations. The electromagnetism-like optimization (EMO) algorithm is an evolutionary technique which emulates the attraction–repulsion mechanism among charges for evolving the individuals of a population. Such technique requires only a small number of evaluations to find the optimum. This paper proposes the use of EMO to search for optimal threshold values by minimizing the cross entropy function while reducing the amount of iterations and function evaluations. The approach is tested on a set of benchmark images to demonstrate that is able to improve the convergence and velocity; additionally, it is compared with similar state-of-the-art optimization approaches.

An efficient method for multilevel color image thresholding using cuckoo search algorithm based on minimum cross entropy

Among various thresholding methods, minimum cross entropy is implemented for its effectiveness and simplicity. Although it is efficient and gives excellent result in case of bi-level thresholding, but its evaluation becomes computationally costly when extended to perform multilevel thresholding owing to the exhaustive search performed for the optimum threshold values. Therefore, in this paper, an efficient multilevel thresholding technique based on cuckoo search algorithm is adopted to render multilevel minimum cross entropy more practical and reduce the complexity. Experiments have been conducted over different color images including natural and satellite images exhibiting low resolution, complex backgrounds and poor illumination. The feasibility and efficiency of proposed approach is investigated through an extensive comparison with multilevel minimum cross entropy based methods that are optimized using artificial bee colony, bacterial foraging optimization, differential evolution, and wind driven optimization. In addition, the proposed approach is compared with thresholding techniques depending on between-class variance (Otsu) method and Tsalli’s entropy function. Experimental results based on qualitative results and different fidelity parameters depicts that the proposed approach selects optimum threshold values more efficiently and accurately as compared to other compared techniques and produces high quality of the segmented images.