Search

The intercity freight trips of heavy trucks are important data for transportation system planning and urban agglomeration management. In recent decades, the extraction of freight trips from GPS data has gradually become the main alternative to traditional surveys. Identifying the trip ends (origin and destination, OD) is the first task in trip extraction. In previous trip end identification methods, some key parameters, such as speed and time thresholds, have mostly been defined on the basis of empirical knowledge, which inevitably lacks universality. Here, we propose a data-driven trip end identification method. First, we define a speed threshold by analyzing the speed distribution of heavy trucks and identify all truck stops from raw GPS data. Second, we define minimum and maximum time thresholds by analyzing the distribution of the dwell times of heavy trucks at stop location and classify truck stops into three types based on these time thresholds. Third, we use highway network GIS data and freight-related points-of-interest (POIs) data to identify valid trip ends from among the three types of truck stops. In this step, we detect POI boundaries to determine whether a heavy truck is stopping at a freight-related location. We further analyze the spatiotemporal characteristics of intercity freight trips of heavy trucks and discuss their potential applications in practice.

Intracity heavy truck freight trips are basic data in city freight system planning and management. In the big data era, massive heavy truck GPS trajectories can be acquired cost effectively in real-time. Identifying freight trip ends (origins and destinations) from heavy truck GPS trajectories is an outstanding problem. Although previous studies proposed a variety of trip end identification methods from different perspectives, these studies subjectively defined key threshold parameters and ignored the complex intracity heavy truck travel characteristics. Here, we propose a data-driven trip end identification method in which the speed threshold for identifying truck stops and the multilevel time thresholds for distinguishing temporary stops and freight trip ends are objectively defined. Moreover, an appropriate time threshold level is dynamically selected by considering the intracity activity patterns of heavy trucks. Furthermore, we use urban road networks and point-of-interest (POI) data to eliminate misidentified trip ends to improve method accuracy. The validation results show that the accuracy of the method we propose is 87.45%. Our method incorporates the impact of the city freight context on truck trajectory characteristics, and its results can reflect the spatial distribution and chain patterns of intracity heavy truck freight trips, which have a wide range of practical applications.

Quantum mechanics is one of the basic theories of modern physics. Here, the famous Schr\"odinger equation and the differential operators representing mechanical quantities in quantum mechanics are derived, just based on the principle that the translation invariance (symmetry) of a system in Hamiltonian mechanics should be preserved in quantum mechanics. Moreover, according to the form of the differential operators, the commutation relation in quantum mechanics between the generalized coordinate and the generalized momentum can be directly obtained. We believe that the results in this paper are very useful for understanding the physical origin of quantum mechanics.

When encountering a dubious diagnostic case, medical instance retrieval can help radiologists make evidence-based diagnoses by finding images containing instances similar to a query case from a large image database. The similarity between the query case and retrieved similar cases is determined by visual features extracted from pathologically abnormal regions. However, the manifestation of these regions often lacks specificity, i.e., different diseases can have the same manifestation, and different manifestations may occur at different stages of the same disease. To combat the manifestation ambiguity in medical instance retrieval, we propose a novel deep framework called Y-Net, encoding images into compact hash-codes generated from convolutional features by feature aggregation. Y-Net can learn highly discriminative convolutional features by unifying the pixel-wise segmentation loss and classification loss. The segmentation loss allows exploring subtle spatial differences for good spatial-discriminability while the classification loss utilizes class-aware semantic information for good semantic-separability. As a result, Y-Net can enhance the visual features in pathologically abnormal regions and suppress the disturbing of the background during model training, which could effectively embed discriminative features into the hash-codes in the retrieval stage. Extensive experiments on two medical image datasets demonstrate that Y-Net can alleviate the ambiguity of pathologically abnormal regions and its retrieval performance outperforms the state-of-the-art method by an average of 9.27\% on the returned list of 10., Comment: 11 pages,8 figures, JBHI Journal

Although it has been surpassed by many subsequent coding standards, JPEG occupies a large share of the storage load of the current data hosting service. To reduce the storage costs, DropBox proposed a lossless secondary compression algorithm, Lepton, to further improve the compression rate of JPEG images. However, the bloated probability models defined by Lepton severely restrict its throughput and energy efficiency. To solve this problem, we construct an efficient access probability-based hash function for the probability models, and then propose a hardware-friendly memory optimization method by combining the proposed hash function and the N-way Set-Associative unit. After that, we design a highly parameterized hardware structure for the probability models and finally implement a power and area efficient Lepton hardware encoder. To the best of our knowledge, this is the first hardware implementation of Lepton. The synthesis result shows that the proposed hardware structure reduces the total area of the probability models by 70.97%. Compared with DropBox's software solution, the throughput and the energy efficiency of the proposed Lepton hardware encoder are increased by 55.25 and 4899 times respectively. In terms of manufacturing cost, the proposed Lepton hardware encoder is also significantly lower than the general-purpose CPU used by DropBox.