Using Fuzzy Clustering of User Perception to Determine the Number of Level-of-Service Categories for Bus Rapid Transit

The concept of level of service (LOS) is meant to reflect user perception of the quality of service provided by a transportation facility or service. Although the LOS of bus rapid transit (BRT) has received considerable attention, the number of levels of service of BRT that a user can perceive still remains unclear. Therefore, in this paper, we address this issue using fuzzy clustering of user perception. User perception is defined as a six-dimension vector of the perceived arrival time, perceived waiting time, bus speed perception, passenger load perception, perceived departure time, and overall perception. A smartphone-based transit travel survey system was developed, with which user perception surveys were conducted in three BRT systems in China. Fuzzy C-Means clustering, improved using a simulated annealing genetic algorithm, was adopted to partition user perception into two to ten clusters. Seven cluster validity indices were used to determine the appropriate number of LOS categories. Our results indicate that users can perceive two to four levels of service.

Mapping Fluvial Landforms with Deep Similarity Learning

<p>Semantic image classification as practised in Earth Observation is poorly suited to mapping fluvial landforms which are often composed of multiple landcover types such as water, riparian vegetation and exposed sediment. Deep learning methods developed in the field of computer vision for the purpose of image classification (ie the attribution of a single label to an image such as cat/dog/etc) are in fact more suited to such landform mapping tasks. Notably, Convolutional Neural Networks (CNN) have excelled at the task of labelling images. However, CNN are notorious for requiring very large training sets that are laborious and costly to assemble. Similarity learning is a sub-field of deep learning and is better known for one-shot and few-shot learning methods. These approaches aim to reduce the need for large training sets by using CNN architectures to compare a single, or few, known examples of an instance to a new image and determining if the new image is similar to the provided examples. Similarity learning rests on the concept of image embeddings which are condensed higher-dimension vector representations of an image generated by a CNN. Ideally, and if a CNN is suitably trained, image embeddings will form clusters according to image classes, even if some of these classes were never used in the initial CNN training.</p><p> </p><p>In this paper, we use similarity learning for the purpose of fluvial landform mapping from Sentinel-2 imagery. We use the True Color Image product with a spatial resolution of 10 meters and begin by manually extracting tiles of 128x128 pixels for 4 classes: non-river, meandering reaches, anastomosing reaches and braiding reaches. We use the DenseNet121 CNN topped with a densely connected layer of 8 nodes which will produce embeddings as 8-dimension vectors. We then train this network with only 3 classes (non-river, meandering and anastomosing) using a categorical cross-entropy loss function. Our first result is that when applied to our image tiles, the embeddings produced by the trained CNN deliver 4 clusters. Despite not being used in the network training, the braiding river reach tiles have produced embeddings that form a distinct cluster. We then use this CNN to perform few-shot learning with a Siamese triplet architecture that will classify a new tile based on only 3 examples of each class. Here we find that tiles from the non-river, meandering and anastomising class were classified with F1 scores of 72%, 87% and 84%, respectively. The braiding river tiles were classified to an F1 score of 80%. Whilst these performances are lesser than the 90%+ performances expected from conventional CNN, the prediction of a new class of objects (braiding reaches) with only 3 samples to 80% F1 is unprecedented in river remote sensing. We will conclude the paper by extending the method to mapping fluvial landforms on entire Sentinel-2 tiles and we will show how we can use advanced cluster analyses of image embeddings to identify landform classes in an image without making a priori decisions on the classes that are present in the image.</p>

JPEG Lifting Algorithm Based on Adaptive Block Compressed Sensing

This paper proposes a JPEG lifting algorithm based on adaptive block compressed sensing (ABCS), which solves the fusion between the ABCS algorithm for 1-dimension vector data processing and the JPEG compression algorithm for 2-dimension image data processing and improves the compression rate of the same quality image in comparison with the existing JPEG-like image compression algorithms. Specifically, mean information entropy and multifeature saliency indexes are used to provide a basis for adaptive blocking and observing, respectively, joint model and curve fitting are adopted for bit rate control, and a noise analysis model is introduced to improve the antinoise capability of the current JPEG decoding algorithm. Experimental results show that the proposed method has good performance of fidelity and antinoise, especially at a medium compression ratio.

Birational geometry of symplectic quotient singularities

For a finite subgroup $$\Gamma \subset \mathrm {SL}(2,\mathbb {C})$$ Γ ⊂ SL ( 2 , C ) and for $$n\ge 1$$ n ≥ 1 , we use variation of GIT quotient for Nakajima quiver varieties to study the birational geometry of the Hilbert scheme of n points on the minimal resolution S of the Kleinian singularity $$\mathbb {C}^2/\Gamma $$ C 2 / Γ . It is well known that $$X:={{\,\mathrm{{\mathrm {Hilb}}}\,}}^{[n]}(S)$$ X : = Hilb [ n ] ( S ) is a projective, crepant resolution of the symplectic singularity $$\mathbb {C}^{2n}/\Gamma _n$$ C 2 n / Γ n , where $$\Gamma _n=\Gamma \wr \mathfrak {S}_n$$ Γ n = Γ ≀ S n is the wreath product. We prove that every projective, crepant resolution of $$\mathbb {C}^{2n}/\Gamma _n$$ C 2 n / Γ n can be realised as the fine moduli space of $$\theta $$ θ -stable $$\Pi $$ Π -modules for a fixed dimension vector, where $$\Pi $$ Π is the framed preprojective algebra of $$\Gamma $$ Γ and $$\theta $$ θ is a choice of generic stability condition. Our approach uses the linearisation map from GIT to relate wall crossing in the space of $$\theta $$ θ -stability conditions to birational transformations of X over $$\mathbb {C}^{2n}/\Gamma _n$$ C 2 n / Γ n . As a corollary, we describe completely the ample and movable cones of X over $$\mathbb {C}^{2n}/\Gamma _n$$ C 2 n / Γ n , and show that the Mori chamber decomposition of the movable cone is determined by an extended Catalan hyperplane arrangement of the ADE root system associated to $$\Gamma $$ Γ by the McKay correspondence. In the appendix, we show that morphisms of quiver varieties induced by variation of GIT quotient are semismall, generalising a result of Nakajima in the case where the quiver variety is smooth.

Integrated System for Classification of Pulmonary Nodules on CT Images

Mortality rate of lung cancer is increasing very day all over the world. Early stage lung nodules detection and proper treatment is solution to reduce the deaths due to lung cancer. In this research work proposed integrated CADe/CADx system segments and classifies lung nodules into benign or malignant. CADe phase segments Well Circumscribed Nodules (WCN), Juxta Vascular Nodules (JVN) and Juxta Pleural Nodules (JPN) of different size in diameter. This part uses algorithms proposed in our previous WCN, JVN and JPN lung nodules segmentation work. CADx performance classification of segmented WCNs, JVNs and JPNs nodules into benign or malignant. In first part of CADx system hybrid features of segmented lung nodules are extracted and features dimension vector is reduced with Linear Discrimination Analysis. Finally, Probabilistic Neural Network uses reduced hybrid features of segmented nodules to classify segmented nodules as benign or malignant. Proposed integrated system achieved high classification accuracy of 94.85 for WCNs, 97.65 for JVNs and 97.96 for JPNs of different size in diameter (nodules diameter< 10mm, nodules diameter >10mm and < 30mm, nodules diameter >30mm and <70mm). For small nodules achieved classification performance values are, accuracy of 94.85, sensitivity of 90 and specificity of 95.85. And nodules of size 10mm to 30mm obtained accuracy, sensitivity and specificity are 97.85, 97.65 and 94.15 respectively.

MATRIKS BAKU UNTUK TRANSFORMASI LINIER PADA RUANG VEKTOR DIMENSI TIGA

The linear transformation is a function relating the vector ke . If , then the transformation is called a linear operator. Several examples of linear operators have been introduced since SMA such as reflexive, rotation, compression and expansion and shear. Apart from being introduced in SMA, these linear operators were also introduced to the linear algebra course. Linear transformations studied at the university level include linear transformation in finite dimension vector spaces . The discussion includes how to determine the standard matrix for reflexive linear transformations, rotation, compression and expansion and given shear. Through the column vectors of reflexive, rotation, compression and expansion and shear, a standard matrix of 2x2 size is formed for the corresponding linear transformation. however, in this study, the authors studied linear transformations in dimensioned vector spaces . The results of this study are if known is a vector space with finite and the standard matrix for reflexivity, rotation, expansion, compression and shear is obtained. Each of these linear transformations is performed on x-axis, y-axis and z-axis on to get column vectors. The column vectors as a result of the linear transformation at form the standard matrix for the corresponding linear transformation in the vector space. The standard matrix for linear transformations in the vector space is obtained by determining reflexivity, rotation, expansion, compression and shear. The process of obtaining a standard matrix for linear transformation is carried out by rewriting the standard basis, determining the column vectors, and rearranging them as the standard matrix for each linear transformation in the vector space

A Network Integration Method for Deciphering the Types of Metabolic Pathway of Chemicals with Heterogeneous Information

Aim and Objective: A metabolic pathway is an important type of biological pathway, which is composed of a series of chemical reactions. It provides essential molecules and energies for living organisms. To date, several metabolic pathways have been uncovered. However, their completeness is still on the way. A number of prediction methods have been built to assign chemicals into certain metabolic pathway, which can further be used to predict novel latent chemicals for a given metabolic pathway. However, they did not make use of chemical properties in a system level to construct prediction models. Method: In this study, we applied a network integration method, which can extract topological features from different chemical networks, representing chemical associations from their different properties, and fused several high-dimension vector representations into a low-dimension vector representation for each chemical. The compact vector representations were fed into the Support Vector Machine (SVM) to construct the prediction model. To tackle the problem that one chemical can participate in more than one pathway type, we construct an SVM-based binary prediction model for each pathway type to determine whether a given chemical can participate in the pathway type. Furthermore, the Synthetic Minority Over-sampling Technique (SMOTE) was adopted to weaken the influence of imbalanced dataset. Results and Conclusion: Each binary model gave a quite good performance and was superior to the classic prediction model, indicating that the proposed models can be useful tools for integrating heterogeneous information to assign chemicals into correct metabolic pathways.