Similarity Measure Method Research Articles

Unknown defect detection for printed circuit board based on multi‐scale deep similarity measure method

Defect detection with high precision is of great significance for printed circuit board (PCB) fabrication. Due to the lack of priori knowledge of categories and shape features, detection of unknown defects faces greater challenges than that of common defects. Inspired by similarity measurement, this study proposes a multi-layer deep feature fusion method to calculate the similarity between template and defective circuit board. Compared with conventional methods which divide the whole detection into two independent parts of hand-designed features and similarity measurement, the authors end-to-end model is designed to combine these two parts for joint optimisation. First, the Siamese network is utilised as their backbone architecture for feature extraction of pairwise images. And then the spatial pyramid pooling network is incorporated into the feature maps of each convolutional module to fuse the multi-scale feature vectors. Finally, the discriminative feature embedding and similarity metric are obtained by using the contrastive loss during the training process. Experimental results show that the proposed model has better performance in detecting and locating unknown defects in bare PCB images than traditional similarity measurement methods. Moreover, our method is promising for further improvement of defect detection with less training image pairs and more accurate detection results.

Структурування критерiального простору за кутовою мiрою подiбностi

Multicriteria decision-making is a particularly hard complex of tasks for a person’s information processing system. As a rule, the more the problem model is constructed and reflects the real problem or task that caused it, the more criteria it has to take into account. With this dimension, classical methods of mathematical programming are ineffective. This necessitates the development of specific methods and approaches designed to structure the criterion space of large dimension problems.This paper describes a fuzzy binary relation and its belonging function that determine the angular measure of similarity of efficiency criteria. It characterizes the degree of similarity between the vector gradients of the objective functions of the efficiency criteria between them. The one-tier clustering method was modified based on fuzzy binary relations to use the angular similarity measure. This allowed clustering of the criterion space into conical clusters on the basis of similarity - a consistent strong link between the performance criteria. The complex approach to structuring the criterion space of vector linear programming problems is presented. On the basis of the proposed mathematical apparatus, software was developed that implements clustering with conical clusters. Conducting hands-on experiments has shown its effectiveness in solving certain classes of application tasks. This work is an evolution of the direction of structuring the set of efficiency criteria for a class of multicriteria linear programming problems with a large dimensional criterion space inconditions where it is difficult or impossible to group, compare or order partial criteria, preferably for the decision maker. Prospective research is to develop a proposed clustering approach which is based on method of fuzzy binary angular similarity measures for solving other classes of applied problems.

Improved k-means and spectral matching for hyperspectral mineral mapping

Mineral mapping is an important step for the development and utilization of mineral resources. The emergence of remote sensing technology, especially hyperspectral imagery, has paved a new approach to geological mapping. The k-means clustering algorithm is a classical approach to classifying hyperspectral imagery, but the influence of mixed pixels and noise mean that it usually has poor mineral mapping accuracy. In this study, the mapping accuracy of the k-means algorithm was improved in three ways: similarity measurement methods that are insensitive to dimensions are used instead of the Euclidean distance for clustering; the spectral absorption features of minerals are enhanced; and the mineral mapping results are combined as the number of cluster centers (K) is incremented from 1. The improved algorithm is used with combined spectral matching to match the clustering results with a spectral library. A case study on Cuprite, Nevada, demonstrated that the improved k-means algorithm can identify most minerals with the kappa value of over 0.8, which is 46% and 15% higher than the traditional k-means and spectral matching technology. New mineral types are more likely to be found with increasing K. When K is much greater than the number of mineral types, the accuracy is improved, and the mineral mapping results are independent of the similarity measurement method. The improved k-means algorithm can also effectively remove speckle noise from the mineral mapping results and be used to identify other objects.

A new approach to similarity measure for generalized trapezoidal fuzzy numbers and its application to fuzzy risk analysis

A new similarity measure has been proposed in this paper for generalized trapezoidal fuzzy numbers (GTrFNs). Here, the proposed similarity measure has been devised based on exponent distance, area, perimeter and height of a GTrFN. Most of the researchers described the similarity measure between two GTrFNs whose components belong to [0, 1] only. But the components of GTrFNs should be any real numbers belonged to R. For this reason, in this paper a similarity measure technique has been framed newly on GTrFN whose components are any real number which is the most important consideration here. Depending on the proposed method of similarity measure, some essential properties have been illustrated in this paper. Also, this method has been compared with some existing techniques of similarity measure taking twenty five different sets of GTrFNs. Henceforth, it is obtained that our proposed technique is better than other existing techniques. Finally, the proposed similarity measure has been applied in fuzzy risk analysis in a production system which has been described with numerical illustration.

A Proposed Similarity Measure for Text-Classification

A Similarity measure is a main process in the text processing method. we have proposed a new method of similarity measure which improved the performance of the K-NN method. The proposed measure extended the accuracy of the text classification method. we have implemented proposed method with Amazon dataset and we observed the effectiveness of proposed similarity measure is increase the accuracy of the similarity between set of documents in a corpus. The end result display performance achieved by way of the proposed measure is higher than that obtained by others.

DLCSS: A new similarity measure for time series data mining

The Longest Common Subsequence (LCSS) is considered as a classic problem in computer science. In most studies related to time series data mining, LCSS had been mentioned as the best and the most usable similarity measurement method. The results of time series data mining under LCSS strongly depend on the similarity threshold, because the similarity measurement approach in LCSS is a zero–one approach. Since there is no knowledge about the data, and it is very difficult to determine the right amount of similarity threshold, using LCSS can actually lead to poor results. In this research, a new similarity measurement method named Developed Longest Common Subsequence (DLCSS) has been suggested for time series data mining based on LCSS. In DLCSS, by defining two similarity thresholds and determining their values, LCSS’ shortcoming was eliminated. The performance of DLCSS was compared with performance of LCSS and Dynamic Time Warping (DTW) using 1-Nearest neighbor and k-medoids clustering techniques. This evaluation was carried out on 63 time series datasets of UCR collection. Using these results, it could be claimed that the 1-NN accuracy and clustering accuracy under DLCSS is better than that of under LCSS and DTW with at least 99.5% and 99% confidence, respectively. Also, DLCSS has better effect in correctly predicting the number of clusters compared to LCSS and DTW. In addition, the effect of DLCSS in determining the better cluster representatives is greater than that of under LCSS and DTW with at least 99.95% confidence.

A sample entropy inspired affinity propagation method for bearing fault signal classification

For bearing fault signal classification, the performance of the affinity propagation (AP) method is limited by its way of measuring similarity. In this paper, we have proposed an improved version for the AP method, which is termed as IMFSE-AP. The key point of the proposed IMFSE-AP method is the improvement on the way of measuring signal similarity. Instead of the Euclidean distance, a novel method based on the sample entropy of IMFs (IMFSE) is proposed, which can measure signal similarities better with respect to data complexity. In the proposed similarity measurement method, signals are first decomposed into IMFs by the EEMD method. Then the distances of sample entropy vectors of IMFs are computed to evaluate similarities between signals. Numerical experiments conducted on synthetic signals and real bearing fault signals illustrate the good performance of the proposed method.

A multi-level similarity measure for the retrieval of the common CT imaging signs of lung diseases.

The common CT imaging signs of lung diseases (CISLs) which frequently appear in lung CT images are widely used in the diagnosis of lung diseases. Computer-aided diagnosis (CAD) based on the CISLs can improve radiologists' performance in the diagnosis of lung diseases. Since similarity measure is important for CAD, we propose a multi-level method to measure the similarity between the CISLs. The CISLs are characterized in the low-level visual scale, mid-level attribute scale, and high-level semantic scale, for a rich representation. The similarity at multiple levels is calculated and combined in a weighted sum form as the final similarity. The proposed multi-level similarity method is capable of computing the level-specific similarity and optimal cross-level complementary similarity. The effectiveness of the proposed similarity measure method is evaluated on a dataset of 511 lung CT images from clinical patients for CISLs retrieval. It can achieve about 80% precision and take only 3.6ms for the retrieval process. The extensive comparative evaluations on the same datasets are conducted to validate the advantages on retrieval performance of our multi-level similarity measure over the single-level measure and the two-level similarity methods. The proposed method can have wide applications in radiology and decision support. Graphical abstract.

A Text Similarity Measurement Based on Semantic Fingerprint of Characteristic Phrases

Text similarity measurements are the basis for measuring the degree of matching between two or more texts. Traditional large-scale similarity detection methods based on a digital fingerprint have the advantage of high detection speed, which are only suitable for accurate detection. We propose a method of Chinese text similarity measurement based on feature phrase semantics. Natural language processing (NLP) technology is used to pre-process text and extract the keywords by the Term frequency-Inverse document frequency (TF-IDF) model and further screen out the feature words. We get the exact meaning of a word and semantic similarities between words and a HowNet semantic dictionary. We substitute concepts to get the feature phrases and generate a semantic fingerprint and calculate similarity. The experimental results indicate that the method proposed is superior in similarity detection in terms of its accuracy rate, recall rate, and F-value to the traditional and digital fingerprinting method.

RGB-D image segmentation using superpixel and multi-feature fusion graph theory

It is difficult to obtain accurate segmentation results for a color map when there are shadows, low-contrast edges, or blurred regions in the image. The depth discontinuity of the image provides useful information for the identification of object boundaries. In this paper, a color and depth image (RGB-D) image segmentation method based on superpixels and multi-feature fusion graph theory is proposed. The method consists of two stages: (1) superpixel segmentation stage and (2) graph-based superpixel merge stage. Color information and depth information are combined to establish a Euclidean distance metric for the pixels, and superpixels are obtained through iterative clustering. A multi-feature fusion adjacent superpixel similarity measurement method based on the RGB-D data is proposed. A graph-based energy function is established, and a label cost is introduced in the energy function to eliminate redundant labels. We carried out many experiments on the RGB-D image database, and the results show that our superpixel segmentation method and graph-based superpixel merging method not only have higher segmentation accuracy than the existing methods, but also have advantages in terms of running time and memory consumption.

Exploring Deep Learning for View-Based 3D Model Retrieval

In recent years, view-based 3D model retrieval has become one of the research focuses in the field of computer vision and machine learning. In fact, the 3D model retrieval algorithm consists of feature extraction and similarity measurement, and the robust features play a decisive role in the similarity measurement. Although deep learning has achieved comprehensive success in the field of computer vision, deep learning features are used for 3D model retrieval only in a small number of works. To the best of our knowledge, there is no benchmark to evaluate these deep learning features. To tackle this problem, in this work we systematically evaluate the performance of deep learning features in view-based 3D model retrieval on four popular datasets (ETH, NTU60, PSB, and MVRED) by different kinds of similarity measure methods. In detail, the performance of hand-crafted features and deep learning features are compared, and then the robustness of deep learning features is assessed. Finally, the difference between single-view deep learning features and multi-view deep learning features is also evaluated. By quantitatively analyzing the performances on different datasets, it is clear that these deep learning features can consistently outperform all of the hand-crafted features, and they are also more robust than the hand-crafted features when different degrees of noise are added into the image. The exploration of latent relationships among different views in multi-view deep learning network architectures shows that the performance of multi-view deep learning outperforms that of single-view deep learning features with low computational complexity.

A similarity measure method fusing deep feature for mammogram retrieval.

Breast cancer is one of the most important malignant tumors among women causing a serious impact on women's lives and mammography is one the most important methods for breast examination. When diagnosing the breast disease, radiologists sometimes may consult some previous diagnosis cases as a reference. But there are many previous cases and it is important to find which cases are the similar cases, which is a big project costing lots of time. Medical image retrieval can provide objective reference information for doctors to diagnose disease. The method of fusing deep features can improve the retrieval accuracy, which solves the "semantic gap" problem caused by only using content features and location features. A similarity measure method combining deep feature for mammogram retrieval is proposed in this paper. First, the images are pre-processed to extract the low-level features, including content features and location features. Before extracting location features, registration with the standard image is performed. Then, the Convolutional Neural Network, the Stacked Auto-encoder Network, and the Deep Belief Network are built to extract the deep features, which are regarded as high-level features. Next, content similarity and deep similarity are calculated separately using the Euclidean distance between the query image and the dataset images. The location similarity is obtained by calculating the ratio of intersection to union of the mass regions. Finally, content similarity, location similarity, and deep similarity are fused to form the image fusion similarity. According to the similarity, the specified number of the most similar images can be returned. In the experiment, 740 MLO mammograms are used, which are from women in Northeast China. The content similarity, location similarity, and deep similarity are fused by different weight coefficients. When only considering low-level features, the results are better with fusing 60% content feature similarity and 40% lesion location feature similarity. On this basis, CNN deep similarity, DBN deep similarity, and SAE deep similarity are fused separately. The experiments show that when fusing 60% DBN deep feature similarity and 40% low-level feature similarity, the results have obvious advantages. At this time, the precision is 0.745, recall is 0.850, comprehensive evaluation index is 0.794. We propose a similarity measure method fusing deep feature, content feature, and location feature. The retrieval results show that the precision and recall of this method have obvious advantage, compared with the content-based image retrieval and location-based image retrieval.

Time-aspect-sentiment Recommendation Models Based on Novel Similarity Measure Methods

The explosive growth of e-commerce has led to the development of the recommendation system. The recommendation system aims to provide a set of items that meet users’ personalized needs through analyzing users’ consumption records. However, the timeliness of purchasing data and the implicity of feedback data pose severe challenges for the existing recommendation methods. To alleviate these challenges, we exploit the user’s consumption records from the perspectives of user and item, by modeling the data on both item and user level, where the item-level value reflects the grade of item, and the user-level value reflects the user’s purchase intention. In this article, we collect the description information and the reviews of the items from public websites, then adopt sentiment analysis techniques to model the similarities on user level and item level, respectively. In particular, we extend the traditional latent factor model and propose two novel methods— I tem L evel Similarity M atrix F actorization (ILMF) and U ser L evel Similarity M atrix F actorization (ULMF)—by introducing two novel similarity measure methods. In ILMF and ULMF, the consistency between latent factors and explicit aspects is naturally incorporated into learning latent factors of the users and items, such that we can predict the users’ preferences on different items more accurately. Moreover, we propose I tem- U ser L evel Similarity M atrix F actorization (IULMF), which combines these two methods to study their contributions on the final performance. Experimental evaluations on the real datasets show that our methods outperform the baseline approaches in terms of both the precision and NDCG.

Research and Implementation of Object Tracking

Aiming at the problems that the traditional MeanShift has a singleness feature, this paper introduces depth feature based on binocular vision, thus proposes a depth feature based real-time MeanShift tracking algorithm. Firstly, a similarity measurement method based on advanced quadratic-form distance is put forward. Secondly, the self-adaptive adjustment mechanism of feature weight coefficient and updating strategy of object model are improved based on the color and depth features. Finally, the algorithm is simulated by Matlab. Results show that the proposed algorithm performs well in tracking accuracy, robustness.

Multi-Player Tracking for Multi-View Sports Videos with Improved K-Shortest Path Algorithm

Sports analysis has recently attracted increasing research efforts in computer vision. Among them, basketball video analysis is very challenging due to severe occlusions and fast motions. As a typical tracking-by-detection method, k-shortest paths (KSP) tracking framework has been well used for multiple-person tracking. While effective and fast, the neglect of the appearance model would easily lead to identity switches, especially when two or more players are intertwined with each other. This paper addresses this problem by taking the appearance features into account based on the KSP framework. Furthermore, we also introduce a similarity measurement method that can fuse multiple appearance features together. In this paper, we select jersey color and jersey number as two example features. Experiments indicate that about 70% of jersey color and 50% of jersey number over a whole sequence would ensure our proposed method preserve the player identity better than the existing KSP tracking method.

Assessing semantic similarity between concepts: A weighted‐feature‐based approach

SummaryTraditional feature‐based semantic similarity (SS) approaches exploit the Wikipedia features in term of sets. They evaluate the similarity of concepts based on the commonalities among their feature sets. However, these feature‐based approaches treat all the features equally in similarity evaluation. Therefore, they ignore the underlying statistics of the features and consequently lose the essential semantic details about them. One solution is that each feature can be assigned a specific weight using its statistics. This weight will reflect the relative importance of a feature in similarity evaluation. Therefore, in this paper, based on two statistical models, ie, information content and TFIDF, we propose some hybrid semantic similarity measurement methods. Firstly, we propose some new methods called weighting functions to compute the weights of the features and feature sets in Wikipedia. Secondly, based on the weighting functions, we propose some new weighted feature‐based SS approaches for Wikipedia concepts. Thirdly, we evaluate the proposed methods on well‐known benchmarks for English, German, French, and Spanish languages. Finally, we compare the performance of our methods with the traditional feature‐based and some state‐of‐the‐art SS approaches. The experimental evaluation shows that our weighted methods perform better than the traditional feature‐based and some state‐of‐the‐art approaches in similarity evaluation.

Comparative Analysis on Classical Meta-Metric Models for Few-Shot Learning

Few-shot learning are methods and scenarios learned from a small amount of labeled data. While recent meta-metric learning methods have made significant progress, there are still questions about what is the key point of these methods and how they work. To address these problems, in this paper, we evaluate the effects of different parts in classical models. To be specific, we 1) use four typical networks AlexNet, VGG16, GoogLeNet, and ResNet50 to replace the original feature extraction part in the Matching Network, Prototypical Network, and Relation Network, and compare the best results with 17 state-of-the-art meta-metric learning algorithms. 2) fix the feature extraction part of the Matching Network, Prototypical Network and Relation Network, and change the similarity measurement part of each into L1, L2, Cosine. 3) conduct above three models on datasets of different granularity. The experimental results show that for all models evaluated, the addition of non-pretrained networks will make the classification results worse, which shows that it is easy to overfit when using deep networks for few-shot learning. Changes in similarity measurement methods have a significant impact on results, which shows the importance to choose a suitable measurement. Moreover, there are differences in performance on different granularity datasets.

Stochastic Economic Dispatching Strategy of the Active Distribution Network Based on Comprehensive Typical Scenario Set

The increasing penetration of renewable energy resources in the distribution network has posed great uncertainties and challenges for the system security operation. To model various uncertain factors like the wholesale market price and renewable energy generation in the active distribution network (ADN), a similarity measurement method considering the amplitude, volatility and variation trend is proposed. The Latin hypercube sampling method and Graph Pyramid clustering algorithm are adopted to obtain the comprehensive typical scenario set. Furthermore, this study proposes a scenario-based stochastic day-ahead optimal economic dispatch approach based on typical scenario set. The energy trading between the distribution system and the wholesale energy market, various distributed generators, network topology and power flow model are jointly formulated in the proposed operation model. The effectiveness and scalability of the proposed approach are verified using the IEEE 33-bus system. Numerical simulation results under different implementation scenarios indicate that the proposed approach offers a high computational efficiency and promotes the security and economy of the distribution system operation, which has a promising industrial application value.

A New Similarity Computing Model of Collaborative Filtering

Collaborative filtering has become one of the most widely used methods for a variety of commercial recommendations. The key to collaborative filtering is use similarity calculation formula to find similar neighbors or projects. However, most similarity calculation methods only use the user common score and provide bad recommendations. This paper proposes a new similarity measure method, which effectively utilizes the user context information. The new method uses a singularity factor to adjust nonlinear equation and takes into account the user scoring habits. It can improve the accuracy of the prediction. The new method has been tested on the dataset and compared with other algorithms. The results show that the proposed method can improve the recommendation quality.

A novel method of similarity measures of Trapezoidal intuitionistic fuzzy numbers

A novel method of similarity measures of Trapezoidal intuitionistic fuzzy numbers