Optimal Global Threshold Research Articles

Optimization of General Threshold Value for Preprocessing in Plasmodium Parasites Detection

The high mortality rate of malaria makes it a severe disease that spreads throughout all-region by infected female Anopheles mosquitoes, especially in tropical countries. Accurate early malaria detection is one of the ways to reduce the mortality rate. Microscopy-based malaria examinations are still considered the gold standard. Due to numerous large malaria patients with limited parasitologists, an automated detection system is needed as a second opinion to assist parasitologists. This study proposed an optimization method for finding an optimal global threshold value for pre-processing parasite detection. There were three stages of the proposed method. The first is to pre-process digital microscopic images using color channel selection, contrast stretching, and morphological operation. The second is to find the global threshold value using multiple modified Otsu’s. The third is to determine the optimum global threshold value. In the last stage, predicted threshold values are generated using a pattern recognition approach to determine the optimum global threshold value. The proposed method evaluated 468 microscopic images captured from hundreds of thin smear blood slides. The slides are provided by the Department of Parasitology-UGM and the Eijkman Institute for Molecular Biology. The set image contains 691 malaria parasites in all types and life stages of malaria parasites. The proposed method obtained a sensitivity of 99.6 % and the smallest FPs number compared to without the optimization. It indicates that the proposed method has the potential to be implemented in the initial stages of the malaria detection system.

Optimization of General Threshold Value for Preprocessing in Plasmodium Parasites Detection

The high mortality rate of malaria makes it a severe disease that spreads throughout all-region by infected female Anopheles mosquitoes, especially in tropical countries. Accurate early malaria detection is one of the ways to reduce the mortality rate. Microscopy-based malaria examinations are still considered the gold standard. Due to numerous large malaria patients with limited parasitologists, an automated detection system is needed as a second opinion to assist parasitologists. This study proposed an optimization method for finding an optimal global threshold value for pre-processing parasite detection. There were three stages of the proposed method. The first is to pre-process digital microscopic images using color channel selection, contrast stretching, and morphological operation. The second is to find the global threshold value using multiple modified Otsu’s. The third is to determine the optimum global threshold value. In the last stage, predicted threshold values are generated using a pattern recognition approach to determine the optimum global threshold value. The proposed method evaluated 468 microscopic images captured from hundreds of thin smear blood slides. The slides are provided by the Department of Parasitology-UGM and the Eijkman Institute for Molecular Biology. The set image contains 691 malaria parasites in all types and life stages of malaria parasites. The proposed method obtained a sensitivity of 99.6 % and the smallest FPs number compared to without the optimization. It indicates that the proposed method has the potential to be implemented in the initial stages of the malaria detection system.

Entropy Based Segmentation Model for Kidney Stone and Cyst on Ultrasound Image

Segmentation of abnormal masses in kidney images is a tough task. One of the main challenges is the presence of speckle noise, which will restrain the valuable information for the medical practitioners. Hence, the detection and segmentation of the affected regions vary in accuracies. The proposed model includes pre-processing and segmentation of the diseased region. The pre-processing consists of Gaussian filtering and Contrast Limited Adaptive Histogram Equalization (CLHE) to improve the clarity of the images. Further, segmentation has been done based on the entropy of the image and gamma correction has been done to improve the overall brightness of the images. An optimal global threshold value is selected to extract the region of interest and measures the area. The model is analyzed with statistical parameters like Jaccard index and Dice coefficient and compared with the ground truth images. To check the accuracy of the segmentation, relative error is calculated. This framework can be used by radiologists in diagnosing kidney patients

Lane Detection Based on Two-Stage Noise Features Filtering and Clustering

Lane detection is essential for autonomous vehicles, and vision-based lane detection is widely used in the field of intelligent driving cars because of its low cost. Aiming to solve the problem of false detection caused by surrounding vehicles during lane detection, a new multi-lane detection method with strong anti-interference ability is proposed. The main contribution of this work is the utilization of object detection neural network which is applied to remove vehicles from road images as much as possible. Compared with the traditional feature-based detection algorithms, this method can preserve lane line features more accurately and effectively. Firstly, binarization of global optimal threshold method is utilized to extract the basic features of lanes. In order to further remove non-lane line noises and extract lane features as clear as possible, YOLOv4-tiny object detection network is introduced to detect and filter vehicles on the road. Then the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm with denoising function is applied to cluster the different lane features. Finally, an improved RANSAC (Random Sample Consensus) method is used to perform quadratic curve fitting on the clustered features. The experimental results on dataset show that consumed time of a single frame image is only 0.119 second and the precision of the proposed method can reach 94.4%. Compared with other detection algorithms, the proposed algorithm has good real-time performance, stability and robustness in a variety of complex scenarios.

Fast fusion method of virtual image of product shape based on progressive fill line algorithm

In order to overcome the problems of low precision and speed in the existing fast fusion methods of virtual image of product shape, a new fast fusion method of virtual image of product shape based on progressive fill line algorithm is proposed in this paper. The progressive filling line algorithm is used to fill the virtual image holes of product shape, and dibr technology is introduced to reconstruct the image position and restore the image colour. Finally, the fuzzy minimum error threshold algorithm and Ostu algorithm are used to calculate the optimal threshold and global threshold of the product shape virtual image features, so as to realise the rapid fusion of product shape virtual image. The experimental results show that the design method can ensure the fusion accuracy and clarity of virtual image of product shape, and the shortest fusion time is 7 ms.

Performance of a 3D convolutional neural network in the detection of hypoperfusion at CT pulmonary angiography in patients with chronic pulmonary embolism: a feasibility study

BackgroundChronic pulmonary embolism (CPE) is a life-threatening disease easily misdiagnosed on computed tomography. We investigated a three-dimensional convolutional neural network (CNN) algorithm for detecting hypoperfusion in CPE from computed tomography pulmonary angiography (CTPA).MethodsPreoperative CTPA of 25 patients with CPE and 25 without pulmonary embolism were selected. We applied a 48%–12%–40% training-validation-testing split (12 positive and 12 negative CTPA volumes for training, 3 positives and 3 negatives for validation, 10 positives and 10 negatives for testing). The median number of axial images per CTPA was 335 (min–max, 111–570). Expert manual segmentations were used as training and testing targets. The CNN output was compared to a method in which a Hounsfield unit (HU) threshold was used to detect hypoperfusion. Receiver operating characteristic area under the curve (AUC) and Matthew correlation coefficient (MCC) were calculated with their 95% confidence interval (CI).ResultsThe predicted segmentations of CNN showed AUC 0.87 (95% CI 0.82–0.91), those of HU-threshold method 0.79 (95% CI 0.74–0.84). The optimal global threshold values were CNN output probability ≥ 0.37 and ≤ -850 HU. Using these values, MCC was 0.46 (95% CI 0.29–0.59) for CNN and 0.35 (95% CI 0.18–0.48) for HU-threshold method (average difference in MCC in the bootstrap samples 0.11 (95% CI 0.05–0.16). A high CNN prediction probability was a strong predictor of CPE.ConclusionsWe proposed a deep learning method for detecting hypoperfusion in CPE from CTPA. This model may help evaluating disease extent and supporting treatment planning.

Automatic cell counting for phase-contrast microscopic images based on a combination of Otsu and watershed segmentation method.

Cell counting plays a vital role in biomedical researches. However, manual cell counting is time-consuming, laborious, and low efficiency and has a high counting error rate problem. An automatic counting approach for Hela cells of phase-contrast microscopic images is proposed based on the combination of Otsu and watershed segmentation methods to solve the mentioned issues. Firstly, image preprocessing is performed. Secondly, the Otsu method was used to obtain an automatic global optimal threshold for segmentation to achieve batch counting of images. Thirdly, the marker watershed was performed to separate adherent cells and to avoid over-segmentation simultaneously. Finally, cells in phase-contrast microscopic images were counted by detecting the numbers of connected domains in the binary image. Taking the manual counting result as the counting standard and MIS, INC, and ACC are used as evaluation indicators. The experimental results showed that the average values of MIS, INC, and ACC of the proposed method are only 3.31%, 3.49%, and 96.69%, respectively. Additionally, each cell image was counted only takes 0.65 s on averagely. To further test the performance of the proposed method, a comparative experiment was carried out by Image J, and the result shows that the proposed method has a better counting performance with a higher average accuracy of 96.55% to Image J with 93.39%.The proposed method for cell counting is simple, feasible, fast and high accurate, and it can be used as an effective method for cell counting of the phase-contrast microscopic images.

Wavelet method optimised by ant colony algorithm used for extracting stable and unstable signals in intelligent substations

Partial discharge (PD) signals are an important index to evaluate the operation state of intelligent substations. The correct distinction of PD pulse and interference pulse has become a challenging task. Because of the noise and the low signal-to-noise ratio, the stable signals become non-stationary. The selection of a wavelet basis, the selection rule of threshold λ and the design of the threshold function are the key factors affecting the final denoising effect. Therefore, an enhanced ant colony optimisition wavelet (ACOW) algorithm was applied to find the global optimal threshold through the continuous derivative threshold function and the ant colony optimisation (ACO) algorithm. At the same time the efficiency of adaptive search calculation, was also significantly improved. The method of the ACOW algorithm was compared with the soft wavelet method, gradient-based wavelet method and the genetic optimisation wavelet (GOW) method. Using these four methods to denoise four typical signals, different mean square errors (MSE), magnitude errors (ME) and time costs were obtained. Interestingly, the results show that the ACOW method can achieve the minimum MSE and has less time cost. It generates significantly smaller waveform distortion than the other three threshold estimation methods. In addition, the high efficiency and good quality of the output signals are beneficial to the diagnosis of local discharge signals in intelligent substations.

Related Study Based on Otsu Watershed Algorithm and New Squeeze-and-Excitation Networks for Segmentation and Level Classification of Tea Buds

In this study, image segmentation technology is utilized for segmentation of tea leaves and tender buds and deep learning technology is introduced for tea bud classification. Watershed algorithm has good robustness in the field of image segmentation under complex backgrounds, and the key of the algorithm is to determine the image segmentation threshold, which directly affects the accuracy of segmentation. “Maximum Between-Class Variance Method” (Otsu) as a great algorithm that can obtain the global optimal threshold is applied creatively to traditional watershed algorithm in this paper, which we call “Otsu Watershed Algorithm”. Then the structure of the “Squeeze-and-Excitation” (SE) block is adjusted appropriately to improve the feature presentation ability of the network by embedding into several common deep learning models. Extensive experiments demonstrate that this new SE block has superior accuracy and integration capability on challenging dataset and our tea bud dataset.

Super-efficient enhancement algorithm for infrared night vision imaging system

This paper presents new three proposed approaches for enhancement of Infrared (IR) night vision images. The first approach is based on merging gamma correction with the Histogram Matching (HM). The second approach depends on hybrid gamma correction with Contrast Limited Adaptive Histogram Equalization (CLAHE). The third approach is based on a trilateral enhancement that the IR images pass through three stages: segmentation, enhancement and sharpening. In the first stage, the IR image is divided into segments based on Optimum Global Thresholding (OTSU) method. The second stage, which is the heart of the enhancement approach, depends on Additive Wavelet Transform (AWT) to decompose the image into an approximation and details. Homomorphic enhancement is performed on the detail components, while Plateau Histogram Equalization (PHE) is performed on the approximation plane. Then, the image is reconstructed and subjected to a post-processing high pass filter. Average gradient, Sobel edge magnitude, entropy and spectral entropy has been used as quality metrics for assessment the three proposed approaches. It is clear that the third proposed approach gives superior results to the two proposed approaches point view the quality metrics. On the other hand, clear that the third proposed approach takes long computation time in the implementation with respect to the two proposed approaches. The first proposed approach gives better results to the two proposed approaches from the computation time perspective.

RETRACTED ARTICLE: A proposed framework for diagnosis and breast cancer detection

This paper aims to early Breast Cancer (BC) detection by Mammography (MG) established on the production of excellent images and competent interpretation. This paper proposes two algorithms for Object Detection and Diagnosis (ODD) the BC. The first proposal is relied on merging the features of A Trous Algorithm (AT) with Homomorphic Processing (HP)Using Contrast Limited Histogram Equalization (AHUC) following by Segmentation and Feature Extraction (FE) for Classification (ASFC). The ASFC depends on enhancement using AHUC in addition to pre-processing followed by segmentation using Optimum Global Thresholding (OT) and finally FE for the detection or classification task. The second scheme is based segmentation after Hybrid structure HE and Fuzzy model (HEF) and finally FE for Classification (HEFC) The HEFC depends on improvement using HEF and pre-processing followed by segmentation using OT and finally FE for classification task. The performance quality metrics for the suggested techniques are entropy, average gradient, contrast, Sobel edge magnitude, homogeneity, sensitivity, specificity, precision and accuracy.Simulation results prove that the success of both techniques in detecting the BC objects. By comparing the first and the second presented algorithms, it is clear that the second suggested technique gives superior for the ODD the BC.

Implementation face based cancelable multi-biometric system

This paper suggests novel two proposed cancellable biometric realization techniques recognition and template protection. In this paper, the Homomorphic Key (HK) encoding algorithm is utilized for cancelable face system. In the first suggested scheme, the HK algorithm is applied on the face images. The resultant map is encrypted, in order to the second HK utilized in the HK is produced from the image. This approach can be used to develop a frequency domain procedure for making this system for biometric template protection. The second algorithm presents a new approach to detect the features of face images depending the Speeded Up Robust Features (SURF) and Optimum Global Thresholding (OTSU) method. This algorithm is relied on scale space analysis with the number of SURF feature points as the key parameters for classification. Simulation results using evaluation metrics False Positive Rate (FPR), False Negative Rate (FNR), Equal Error Rate (EER), Receiver Operating Characteristic (ROC) and Area under ROC (AROC) prove that the first proposed cancelable biometric technique with the second key are best with comparing the other keys. The obtained results clear that the second suggested technique has sucesseded in detection the features of the skin, eyes, nose, hair, ears images and the face positions.

About quantifying small color differences in digital images

We discuss aspects of the use and possibilities provided by three-color colorimeters or digital cameras in problems of detecting small color differences by computer vision methods. The spectral dependence of the total color differences between pairs of visually indiscernible monochromatic stimuli is experimentally revealed. An experimental setup based on the UM-2 monochromator is created for producing a digital atlas of monochromatic stimuli at 1-nm increments. The atlas serves to test the color gamut and color differentiation of cameras. It is experimentally shown that in the visible spectral range a color difference of 3 units is detected by pairs of stimuli that are unevenly distributed across the spectrum and differ in wavelengths from 1 to 6 nm. The capabilities of computer vision are tested on the examples of identifying additional texts during a technical and forensic examination of documents. A new algorithm is developed for finding and quantitatively characterizing color difference of inserts based on a digital image of the inscription. In the algorithm, the objective analysis of the image is divided into a block of color segmentation and that of color tone and color difference assessment. With such an approach, the color segmentation block performs preprocessing functions, making a border map for the classes with different colors for the subsequent calculations. The Otsu method of optimal global threshold transformation is for the first time applied to a problem of image segmentation by color saturation. The trial of the algorithm confirms its efficiency in the solution of expert tasks.

Cooperative evolutionary classification algorithm for hyperspectral images

Reliable crop maps are a vital source of information for many economic sectors. Hyperspectral images and efficient supervised classification algorithms can help in obtaining these maps. One of the objectives is to use a spectroradiometer as a source of a priori information for training supervised classification algorithms. Although there are many supervised classification algorithms, few can be trained using collected spectral signatures, such as spectral angle mapper (SAM). This algorithm relies on setting the angle threshold values manually to separate different classes, and it is usually a random process. So, the second objective is to use an automated process to optimize the angle threshold values to improve the efficiency of the SAM algorithm. An innovative cooperative classification algorithm for hyperspectral images based on enhancing the supervised SAM algorithm performance using a genetic algorithm (GA) is presented. The improvement is based on selecting global optimal threshold angle values for different classes using GA. The efficiency of the developed cooperative evolutionary algorithm is proved by classifying hyperspectral images to create maps of major crops, such as wheat, potato, and alfalfa. The source of the hyperspectral images is the Compact High-Resolution Imaging Spectrometer onboard of the Proba satellite. The evaluation of the results showed that the new cooperative evolutionary algorithm classified hyperspectral images with the highest accuracy compared to well-known reliable supervised classification algorithms.

Design of Threshold Segmentation Method for Quantum Image

In this paper, a design method of quantum circuits for threshold segmentation of gray-scale image is proposed, which is a quantum counterpart of classical optimum global thresholding using Otsu’s method. First, we design a quantum circuit to obtain the histogram of the pixels distribution. Then, on this basis, we design the quantum circuits of some sub-modules, such as the cumulative histogram, cumulative mean, multiplication, division, the between-class variance and so on. Finally, all the between-class variances corresponding to each gray-level are sorted. At this time, the gray-level corresponding to the maximum of the between-class variance is the Otsu’s threshold. For all the operations involved in the classical Otsu’s threshold selection, we designed the corresponding quantum circuits, analyzed the computational complexity, and verified the correctness of these circuits by simulations on classical computers.

Optimal Global Threshold based on Two Dimension Otsu for Block Size Decision in Intra Prediction of H.264/AVC Coding

The Advanced Video Coding (H.264/AVC) has proved its ability in finding the tradeoff between the compressed bit rate value and the visual quality of video comparing to the others of traditional coding. One of the most encoder stages consuming time is the intra prediction in which different sizes of a block are exhaustively examined for selecting the suitable block size to the best block mode decision. In this paper, an efficient approach is suggested to select the best block size for the intra prediction adaptively to achieve high compression efficiency. The proposed approach exploits the idea of quad tree decomposition for blocks partitioning based on a predefined threshold value. An optimal global threshold value based on two dimension Otsu technique is suggested for the decision of block division in this work. The proposed technique is carried out on different set of videos resolutions with different quantization parameters using Matlab software. The comparison of the proposed approach with the reference JM18.6 video coding is done in terms of bit rate (BR), time saving and peak signal to noise ratio (PSNR). A tangible acceleration on the running time can be accomplished besides improvement in both of visual quality and bit rate with some of QCIF and CIF videos resolution by the proposed technique. The simulation results demonstrate saving in time by average 42% to 68% with CIF and QCIF videos. Concerning the visual quality in terms of Bjontegaard Delta parameters, the PSNR improved in which its value increased from 0.2 to 1.6, while the value of BR reduced from 0.79 to 15.3 respectively with some videos of resolutions QCIF, CIF and 720p. In addition, the performance of the suggested approach with the high resolution videos achieves minor improvement with some of them while has a slightly degradation with others of them.

Automatic Detection of Exudates in Retinal Images using Region-Based, Neighborhood and Block Operation

Exudates detection is one of the research areas attracting great attention of physicians and scientists. A region-based, neighborhood, block operation and optimal global thresholding are proposed as new methods to exudates detection. The exudates are coarse and fine segmentation following preprocessing steps, i.e., color mapping, image contrast enhancement, fuzzy filtering and optic disc localization. To classify the retinal images into non-exudates and exudates, a set of features such as texture, color, size and the edge is extracted. The exudates procedure succeeded in an overall generalization accuracy of 98.62% with 98.18% sensitivity and 98.32% specificity in local databases. Moreover, the results are presented to show the advantage of the proposed method in a public database with an accuracy of 92.14.

Degraded historical document image binarization using local features and support vector machine (SVM)

This paper presents a support vector machine (SVM) based method for degraded historical document image binarization. Given a degraded historical document image, the proposed method first segments the image into w × w regions and implements a local contrast enhancement in each image block. We then use a SVM to select an optimal global threshold for binarization of each image block. Finally, the entire image is further binarized by a locally adaptive thresholding method. The proposed method has been evaluated over the recent Document Image Binarization Competition (DIBCO) datasets. The experimental results show that our proposed method outperforms other state-of-the-art techniques in terms of F-measure, NRM, DRD, and MPM.

A Novel Denoising Method for an Acoustic-Based System through Empirical Mode Decomposition and an Improved Fruit Fly Optimization Algorithm

Generally, the sound signal produced by transmission unit or cutting unit contains abundant information about the working state of a machine. The acoustic-based diagnosis system presents some distinct advantages in some severe conditions particularly due to its unique non-contact measurement and unlimited use at the installation site. However, the original acoustic signal collected from manufacture process is always polluted by various background noises. In order to eliminate noise components from machinery sound effectively, an empirical mode decomposition (EMD) threshold denoising method optimized by an improved fruit fly optimization algorithm (IFOA) is launched in this paper. The acoustic signal was first decomposed by the adaptive EMD to obtain a series of intrinsic mode functions (IMFs). Then, the soft threshold function was applied to shrink the IMF coefficients. While the threshold of each IMF was determined by statistical estimation and empirical value for traditional EMD denoising, the denoising effect was often not desired and time-consuming. To solve these disadvantages, fruit fly optimization algorithm (FOA) was introduced to search global optimal threshold of each IMF. Moreover, to enhance the group diversity during production of the next generation of fruit flies and balance the local and global searching ability, a variation coefficient and a disturbance coefficient was introduced to the basic FOA. Then, a piece of simulated acoustic signal produced by the train was applied to validate the proposed EMD and IFOA threshold denoising (EMD-IFOA). The simulation results, which decreased 35.40% and 18.92% in mean squared error (MSE) and percent root mean square difference (PRD) respectively, and increased 40.36% in signal-to-noise ratio improvement (SNRimp) compared with basic EMD denoising scheme at SNR = 5 dB, illustrated the effectiveness and superiority of the proposed approach. Finally, the proposed EMD-IFOA was conducted on an actual acoustic-based diagnosis system for cutting state recognition of the coal mining shearer to demonstrate the practical effect.

Multibit-Quantization-Based Collaborative Spectrum Sensing Scheme for Cognitive Sensor Networks

In order to overcome the bandwidth limitations of common control channels, we introduce a multibit quantization scheme to preprocess raw measurements in a collaborative spectrum sensing (CoSS) scheme. After quantifying the raw measurement, the cognitive sensors (CSs) send the multibit sensing information to a sink node (SN), which makes a final decision by using a soft decision fusion rule. In order to better understand the effects of quantization parameters on collaborative spectrum sensing in a cognitive sensor network (CSN), we discuss the following two problems: the average error probability minimum problem (AEPMP) and throughput maximum problem. By identifying the number of quantization bits, number of CSs, and global decision threshold, we can discuss some design considerations for the multibit-quantization-based CoSS scheme. A closed-form expression for the optimal global decision threshold is derived in the process of analyzing the AEPMP. Furthermore, an iterative algorithm with low complexity is proposed to find the optimal parameters for maximizing the throughput of a CSN under a target detection probability constraint. Finally, we investigate the impact of optimized parameters on the system performance of a CSN. Our theoretical analysis is verified through numerical simulations.