Adaptive Threshold Selection Method Research Articles

Adaptive Threshold Learning in Frequency Domain for Classification of Breast Cancer Histopathological Images

Breast cancer has become the most common cancer in the world, and biopsy is the most reliable and widely used technique for detecting breast cancer. However, observation of histopathological images is time-consuming and labor-intensive. Currently, CNN has become the mainstream method for breast cancer histopathological image classification research. However, some studies have found that the optical microscope-generated histopathological images have noise, and the output of a well-trained convolutional neural network in image classification tasks can change drastically due to small variations in the input. Therefore, the quality of the image significantly affects the accuracy of the classification. Wavelet transform is a commonly used denoising method, but the selection of the threshold is a difficult problem, and traditional methods are difficult to find the appropriate threshold quickly and accurately. This paper proposes an adaptive threshold selection method that combines threshold selection steps with deep learning methods by using the threshold as a parameter in the CNN model to train. In this way, we associate the threshold with the classification result of the model and find the appropriate value for that image and task by back-propagation in training. The method was experimented on publicly available datasets BreaKHis and BACH. The results in BreaKHis (40x: 94.37%, 100x: 93.85%, 200x: 91.63%, 400x: 93.31%), and BACH (91.25%) demonstrate that our adaptive threshold selection method can improve classification accuracy and is significantly superior to traditional threshold selection methods.

PEAR: Positional-encoded Asynchronous Autoregression for satellite anomaly detection

This paper proposes a Positional-Encoded Asynchronous AutoRegression (PEAR) method for satellite anomaly detection. We empirically observe that a single classification model can hardly detect unknown anomalous situations and neglect the Markov nature of temporal satellite data. To address this, we adopt an autoregressive model to deal with the prediction of unknown anomaly for satellite data. We further propose a non-uniform temporal encoding method for asynchronous data and a median filtering method for more accurate detection. To reduce the effect of outliers, we employ an adaptive threshold selection method to achieve a more robust classification boundary. We test the proposed method on the time series prediction models including LSTM and transformers and we further employ the positional encoding strategy to improve the modeling capabilities of the Transformer model on high-frequency information. Experiments on real satellite data demonstrate that the proposed PEAR method outperforms the baseline method by 55.79%.

Study on a method for suppressing influence of nonuniform surface emissivity of material in eddy current thermography

The nonuniform emissivity of material surface will adversely affect the detection results of the thermal imaging technology. Aiming at the influence of uneven emissivity on the dynamic detection mode of eddy current thermography, a suppression method based on edge detection is proposed in this paper. Since the traditional threshold selection methods are not ideal, a local adaptive threshold selection (LATS) method is applied in this study. This method can automatically obtain the optimal threshold for edge detection to suppress the influence of uneven emissivity. Roberts, Sobel, Prewitt, and Canny operators are tested by using LATS. The results show that Roberts, Sobel, and Prewitt operators can achieve the suppression effect, but the effect of Sobel and Prewitt are similar and better than Roberts.

Voltage sag assessment of large power grid based on branch change influence domain

The change of a large-scale power grid topology leads to the change of node voltage sag assessment results. Thus, it is necessary to re-simulate massive line fault scenarios. However, the simulation time of the existing assessment methods is too long to ensure the practicability of the project. To solve the problems above, we propose a voltage sag assessment method for a large-scale power grid based on the branch change influence domain. Firstly, the branch change influence index is constructed to quantify the influence law of the branch change on the node voltage sag level. Secondly, based on the different influence degrees of each region, an adaptive electrical distance threshold selection method is proposed and then the range of the branch change influence domain is divided in order to construct and simulate the fault scene in the branch change influence domain. Finally, the historical simulation results are updated by using the simulation results of the branch change influence domain fault, and the severity of the node voltage sag after the topology change is calculated. Through the example of the IEEE118 node system and the actual grid structure of a city in Fujian, it has been proved that the proposed method can effectively reduce the fault simulation time by up to 96% and improve the efficiency of the voltage sag assessment of a large-scale power grid.

Research on Moving Object Detection Based On Probability Statistics Adaptive Background Model of VR Tech

The detection of moving objects often has a high value utilization demand in real life, but the current detection methods of moving objects still have some problems and shortcomings, which need to be further optimized and ameliorated. On account of this, this paper first analyses the common methods of moving object detection, then studies the probability statistics adaptive background threshold selection method on account of VR, and finally gives the construction strategy of background model on account of pixel statistics and probability.

Real-Time Road Curb and Lane Detection for Autonomous Driving Using LiDAR Point Clouds

The commercialization of automated driving vehicles promotes the development of safer and more efficient autonomous driving technologies including lane marking detection strategy, which is considered to be the most promising feature in environmental perception technology. To reduce the tradeoff between time consumption and detection precision, we propose a real-time lane marking detection method by using LiDAR point clouds directly. A constrained RANSAC algorithm is applied to select the regions of interest and filter the background data. Further, a road curb detection method based on the segment point density is also proposed to classify the road points and curb points. Finally, an adaptive threshold selection method is proposed to identify lane markings. In this investigation, five datasets are collected from different driving conditions that include the straight road, curved road, and uphill, to test the proposed method. The proposed method is evaluated under different performance metrics such as Precision, Recall, Dice, Jaccard as well as the average detection distance and computation time for the five datasets. The quantitative results show the efficiency and feasibility of this proposed method.

Efficient match pair selection for oblique UAV images based on adaptive vocabulary tree

The primary contribution of this paper is an efficient match pair selection method for oblique unmanned aerial vehicle (UAV) images. First, high overlap degrees and spatial resolutions cause image and feature redundancies in vocabulary tree building and image indexing. To cope with this issue, an image selection strategy and a feature selection strategy are designed to decrease the total number of features. Second, by analysing the distribution of the similarity scores, an adaptive threshold selection method is implemented to determine the number of candidate match pairs for each query image, and it avoids the disadvantages of the fixed number and fixed proportion methods. Then, an algorithm, termed AVT-Expansion, is proposed for the match pair selection and simplification where the initial match pairs are first selected by using the adaptive vocabulary tree (AVT). To simplify the initial match pairs, the AVT method is integrated with our previous MST-Expansion algorithm, which is used to extract a match graph by analysing the image topological connection network. Finally, the proposed method is verified using three UAV datasets captured with different oblique multi-camera systems. Experimental results demonstrate that the efficiency of the vocabulary tree building is improved, with speed-up ratios ranging from 14 to 16, and that high image retrieval precision values are obtained for the three datasets. For match pair selection of oblique UAV images, the proposed method is an efficient solution.

Coupling characteristic analysis and a fault Pole detection Scheme for single-circuit and double-circuit HVDC transmission lines

The mis-operation and protection issues associated with non-fault poles caused by electromagnetic coupling effects threaten the security and stability of HVDC power systems. Based on the distributed parameter model, the coupling characteristics of single-circuit HVDC transmission lines and double-circuit HVDC transmission lines are studied in this paper. The decoupling method of double-circuit HVDC transmission lines that are structurally unbalanced is discussed. Based on waveform and frequency characteristic analyses of electrical quantities considering coupling effects, a fault pole detection scheme is proposed. A rapid fault pole detection criterion is formulated based on the backward travelling wave, and an adaptive threshold selection method is proposed. A reliable fault pole detection criterion is formulated using the low-frequency component voltage, and a short-window fast-filtering method is proposed. The travelling wave-based fault pole detection method can be rapidly applied using only the electrical quantities of the protected line, and the sensitivity and reliability of the method are enhanced by using adaptive thresholds. Based on the frequency characteristics of the coupling effects, the fault pole detection method can be reliably applied using only the electrical quantities of the protected pole. Simulations demonstrate the feasibility and validity of the fault pole detection scheme.

Adaptive Edge Detection Algorithm Based on Improved Grey Prediction Model

Existing edge detection algorithms suffer from inefficient edge localization, noise sensitivity, and/or relatively poor automatic detection capability. Contemporary edge detection algorithms can be improved by targeting these problems to help bolster their performance. Grey system theory can be used to resolve the small data and poor information issues in the local information of uncertain systems. An automatic edge detection algorithm was developed in this study based on a grey prediction model to remedy these problems. Noise characteristics in grey images are used to deploy a noise-filtering algorithm based on local features. A mask with twenty-four edge direction information points (345°) was established based on edge line texture features. By compressing the amplitude of the sequence, the randomly oscillated grey prediction sequence can be converted into a smooth, new sequence. The discrete grey model (1,1) (DGM(1,1)) was established based on this new grey prediction sequence to obtain the grey prediction maximum value. A grey prediction image with enhanced edges was obtained by replacing the pixel value in the original image with the maximum grey prediction value. A grey prediction subtraction image with edges separated from non-edge points was also obtained by subtracting the original image from the grey prediction image. The optimal separation threshold in the grey prediction subtraction image can be determined via the global adaptive threshold selection method. The neighborhood search method was then deployed to remove stray points and burrs from the image after the target was separated from the background, creating the final edge image. Experiments were performed on a computer-simulated phantom to find that both the subjective visual effects and objective evaluation criteria are better under the proposed method than several other competitive methods. The proposed edge detection algorithm shows excellent edge detection ability and is highly robust to noise, though the grey prediction model needs further improvement to optimize the run time.

A Dark Target Detection Method Based on the Adjacency Effect: A Case Study on Crack Detection.

Dark target detection is important for engineering applications but the existing methods do not consider the imaging environment of dark targets, such as the adjacency effect. The adjacency effect will affect the quantitative applications of remote sensing, especially for high contrast images and images with ever-increasing resolution. Further, most studies have focused on how to eliminate the adjacency effect and there is almost no research about the application of the adjacency effect. However, the adjacency effect leads to some unique characteristics for the dark target surrounded by a bright background. This paper utilizes these characteristics to assist in the detection of the dark object, and the low-high threshold detection strategy and the adaptive threshold selection method under the assumption of Gaussian distribution are designed. Meanwhile, preliminary case experiments are carried out on the crack detection of concrete slope protection. Finally, the experiment results show that it is feasible to utilize the adjacency effect for dark target detection.

An efficient computerized decision support system for the analysis and 3D visualization of brain tumor

The quality of health services provided by medical centers varies widely, and there is often a large gap between the optimal standard of services when judged based on the locality of patients (rural or urban environments). This quality gap can have serious health consequences and major implications for patient’s timely and correct treatment. These deficiencies can manifest, for example, as a lack of quality services, misdiagnosis, medication errors, and unavailability of trained professionals. In medical imaging, MRI analysis assists radiologists and surgeons in developing patient treatment plans. Accurate segmentation of anomalous tissues and its correct 3D visualization plays an important role inappropriate treatment. In this context, we aim to develop an intelligent computer-aided diagnostic system focusing on human brain MRI analysis. We present brain tumor detection, segmentation, and its 3D visualization system, providing quality clinical services, regardless of geographical location, and level of expertise of medical specialists. In this research, brain magnetic resonance (MR) images are segmented using a semi-automatic and adaptive threshold selection method. After segmentation, the tumor is classified into malignant and benign based on a bag of words (BoW) driven robust support vector machine (SVM) classification model. The BoW feature extraction method is further amplified via speeded up robust features (SURF) incorporating its procedure of interest point selection. Finally, 3D visualization of the brain and tumor is achieved using volume marching cube algorithm which is used for rendering medical data. The effectiveness of the proposed system is verified over a dataset collected from 30 patients and achieved 99% accuracy. A subjective comparative analysis is also carried out between the proposed method and two state-of-the-art tools ITK-SNAP and 3D-Doctor. Experimental results indicate that the proposed system performed better than existing systems and assists radiologist determining the size, shape, and location of the tumor in the human brain.

Ellipse-fitting algorithm and adaptive threshold to eliminate outliers

Terrestrial laser scanning is widely applied in many fields owing to its characteristic of rapid acquisition of massive 3D point data. It provides a new way to obtain the cross-section data of metro tunnels for deformation analysis. However, the data contain many outliers, such as pipe and bolt holes, and manual filtering of unwanted points is relatively onerous. Therefore, an ellipse-fitting algorithm based on residual p-norm minimum is proposed to deal with the outliers. Then, an adaptive threshold selection method is introduced for outlier elimination. The remaining valid data are utilised to calculate the deformation after data processing. The experiments validate that the p-norm minimum is more robust than the least-squares algorithm, and the application of an adaptive threshold allows the algorithm to clearly distinguish the outliers. This research provides a reference for the monitoring of subway tunnel deformation.

Positive–negative corresponding normalized ghost imaging based on an adaptive threshold

Ghost imaging (GI) technology has attracted increasing attention as a new imaging technique in recent years. However, the signal-to-noise ratio (SNR) of GI with pseudo-thermal light needs to be improved before it meets engineering application demands. We therefore propose a new scheme called positive–negative correspondence normalized GI based on an adaptive threshold (PCNGI-AT) to achieve a good performance with less amount of data. In this work, we use both the advantages of normalized GI (NGI) and positive–negative correspondence GI (P–NCGI). The correctness and feasibility of the scheme were proved in theory before we designed an adaptive threshold selection method, in which the parameter of object signal selection conditions is replaced by the normalizing value. The simulation and experimental results reveal that the SNR of the proposed scheme is better than that of time-correspondence differential GI (TCDGI), avoiding the calculation of the matrix of correlation and reducing the amount of data used. The method proposed will make GI far more practical in engineering applications.

An Intelligent HARQ Scheme for HSDPA

In this paper, we propose an intelligent hybrid automatic repeat request (iHARQ) scheme for high-speed downlink packet access (HSDPA) systems. The challenge in the hybrid automatic repeat request (HARQ) control problem of HSDPA is how to choose an appropriate modulation and coding scheme (MCS) for initial transmission in the situation wherein the channel quality indication (CQI) has report delay. The iHARQ scheme will determine the suitable MCS to maximize the system throughput and guarantee the block error rate requirement in such an uncertain environment. By modeling the HARQ behavior as a Markov decision process, we adopt fuzzy logics to determine an appropriate MCS for each initial packet transmission. In addition, a Q-learning algorithm is utilized to update the fuzzy rule base, according to well-designed reinforcement signals fed back from the HSDPA system, such that the iHARQ scheme can adapt to the delayed CQI. Simulation results show that, compared with a conventional adaptive threshold selection method, the proposed iHARQ scheme increases the system throughput by up to 75.2%.

Application of Image Processing in Quality Detection of PS Plate

In offset printing system, presensitized plate (PS plate) is one of the key factors that influences the quality of printing products. In this paper, we suggest an application of image processing techniques to improve the accuracy of quality detection of PS plate. We use median filtering to get rid of noises and grayscale stretch to enhance image. An improved adaptive threshold selection method for defect segmentation is proposed based on the characteristic of grads-histogram of PS plate image, and a fast scanning algorithm is put forward to extract features of defects for classification and recognition. The experimental results show that the above methods can obtain the accuracy of 96.7% for defect detection of PS plate and the minimum detectable diameter of defects is 0.2mm, which has reached the enterprise standard for quality detection.

A Robust and Efficient Video Moving Object Detection and Tracking Algorithm

摘要: 提出了一种视频运动目标的快速检测和稳定跟踪算法. 目标检测使用减背景法, 用均值法构造背景图像, 提出一种基于熵能和广义高斯分布的局部自适应阈值选取算法, 可有效克服噪声的影响. 采用基于特征匹配的目标跟踪方法, 提出一种LICS (Logarithm illuminance contrast statistic)特征, 该特征能够更加充分有效地表征目标, 可在光照和目标姿态变化的情况下实现刚体目标的稳定跟踪. 使用Kalman滤波限制搜索匹配范围以减小计算量. 用目标子区域匹配的方法解决目标相互遮挡时的跟踪问题. 实验结果表明, 该算法在运动目标检测效果、跟踪稳定性和运行时间方面都有良好的性能. 关键词: 目标检测与跟踪 / 熵能 / 广义高斯模型 / 自适应阈值

A genetic-based adaptive threshold selection method for dynamic path tree structured vector quantization

This paper presents an improvement method for enhancing the encoding time complexity of the dynamic path tree structured vector quantization (DPTSVQ) based on the same image quality. We call it the genetic-based adaptive threshold selection method (GATSM). DPTSVQ has successfully solved the disadvantage of the multi-path TSVQ. DPTSVQ uses a critical function and a fixed threshold to judge whether the number of search paths can be increased. However, in some cases, the fixed threshold scheme also brings the problem of increasing the encoding time. We thus propose GATSM to solve this problem by using a set of images to train the thresholds for adapting their real practical need. Our experimental results show that the encoding time complexity of GATSM is superior to DPTSVQ based on the same image quality. In addition, we compare the image quality of GATSM with the encoding algorithm with fast comparison (EAWFC) based on the same encoding time. Comparison results show that GATSM provides better image quality than that of EAWFC.