High-Speed Railway Clearance Intrusion Detection with Improved SSD Network

With the rapid development of high-speed railways, any objects intruding railway clearance will do great threat to railway operations. Accurate and effective intrusion detection is very important. An original Single Shot multibox Detector (SSD) can be used to detect intruding objects except small ones. In this paper, high-level features are deconvolved to low-level and fused with original low-level features to enhance their semantic information. By this way, the mean average precision (mAP) of the improved SSD algorithm is increased. In order to decrease the parameters of the improved SSD network, the L1 norm of convolution kernel is used to prune the network. Under this criterion, both the model size and calculation load are greatly reduced within the permitted precision loss. Experiments show that the mAP of our method on PASCAL VOC public dataset and our railway datasets have increased by 2.52% and 4.74% respectively, when compared to the original SSD. With our method, the elapsed time of each frame is only 31 ms on GeForce GTX1060.

Download Full-text

Dense-RefineDet for Traffic Sign Detection and Classification

Sensors ◽

10.3390/s20226570 ◽

2020 ◽

Vol 20 (22) ◽

pp. 6570

Author(s):

Chang Sun ◽

Yibo Ai ◽

Sheng Wang ◽

Weidong Zhang

Keyword(s):

High Speed ◽

Large Scale ◽

Design Method ◽

Contextual Information ◽

Real Life ◽

Single Shot ◽

Traffic Sign ◽

Traffic Signs ◽

High Level ◽

Anchor Design

Detecting and classifying real-life small traffic signs from large input images is difficult due to their occupying fewer pixels relative to larger targets. To address this challenge, we proposed a deep-learning-based model (Dense-RefineDet) that applies a single-shot, object-detection framework (RefineDet) to maintain a suitable accuracy–speed trade-off. We constructed a dense connection-related transfer-connection block to combine high-level feature layers with low-level feature layers to optimize the use of the higher layers to obtain additional contextual information. Additionally, we presented an anchor-design method to provide suitable anchors for detecting small traffic signs. Experiments using the Tsinghua-Tencent 100K dataset demonstrated that Dense-RefineDet achieved competitive accuracy at high-speed detection (0.13 s/frame) of small-, medium-, and large-scale traffic signs (recall: 84.3%, 95.2%, and 92.6%; precision: 83.9%, 95.6%, and 94.0%). Moreover, experiments using the Caltech pedestrian dataset indicated that the miss rate of Dense-RefineDet was 54.03% (pedestrian height > 20 pixels), which outperformed other state-of-the-art methods.

Download Full-text

PENERAPAN DATA MINING DALAM MENGELOMPOKAN PRODUKSI JAGUNG MENURUT PROVINSI MENGGUNAKAN ALGORITMA K-MEANS

KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer) ◽

10.30865/komik.v3i1.1681 ◽

2019 ◽

Vol 3 (1) ◽

Author(s):

Nanda Erlangga ◽

Solikhun Solikhun ◽

Irawan Irawan

Keyword(s):

Data Mining ◽

Clustering Algorithm ◽

Rapid Development ◽

Domestic Market ◽

Corn Production ◽

Low Level ◽

High Level ◽

Level Cluster ◽

Central Statistics

Corn needs are currently experiencing a fairly rapid development can be seen in terms of the domestic market, here researchers want to increase the productivity and quality of corn production. The data that will be used is the data from the Central Statistics Agency. The method in this study is the K-means clustering algorithm and the application used is Rapidminer which will be grouped into 2 clustering, namely high and low. The results of this study are 2 high level cluster provinces, 32 low level cluster provincesKeywords: Corn, Data mining, K-means Clustering c

Download Full-text

Automatic Detection of Track and Fields in China from High-Resolution Satellite Images Using Multi-Scale-Fused Single Shot MultiBox Detector

Remote Sensing ◽

10.3390/rs11111377 ◽

2019 ◽

Vol 11 (11) ◽

pp. 1377 ◽

Cited By ~ 1

Author(s):

Zhengchao Chen ◽

Kaixuan Lu ◽

Lianru Gao ◽

Baipeng Li ◽

Jianwei Gao ◽

...

Keyword(s):

Satellite Images ◽

Automatic Detection ◽

Image Resolution ◽

Track And Field ◽

Single Shot ◽

Feature Map ◽

Low Level ◽

Multi Scale ◽

High Resolution Satellite Images ◽

High Level

Object detection is facing various challenges as an important aspect in the field of remote sensing—especially in large scenes due to the increase of satellite image resolution and the complexity of land covers. Because of the diversity of the appearance of track and fields, the complexity of the background and the variety between satellite images, even superior deep learning methods have difficulty extracting accurate characteristics of track and field from large complex scenes, such as the whole of China. Taking track and field as a study case, we propose a stable and accurate method for target detection. Firstly, we add the “deconvolution” and “concat” module to the structure of the original Single Shot MultiBox Detector (SSD), where Visual Geometry Group 16 (VGG16) is served as a basic network, followed by multiple convolution layers. The two modules are used to sample the high-level feature map and connect it with the low-level feature map to form a new network structure multi-scale-fused SSD (abbreviated as MSF_SSD). MSF-SSD can enrich the semantic information of the low-level feature, which is especially effective for small targets in large scenes. In addition, a large number of track and fields are collected as samples for the whole China and a series of parameters are designed to optimize the MSF_SSD network through the deep analysis of sample characteristics. Finally, by using MSF_SSD network, we achieve the rapid and automatic detection of meter-level track and fields in the country for the first time. The proposed MSF_SSD model achieves 97.9% mean average precision (mAP) on validation set which is superior to the 88.4% mAP of the original SSD. Apart from this, the model can achieve an accuracy of 94.3% while keeping the recall rate in a high level (98.8%) in the nationally distributed test set, outperforming the original SSD method.

Download Full-text

A Compact, Bone-Attached Robot for Mastoidectomy

Journal of Medical Devices ◽

10.1115/1.4030083 ◽

2015 ◽

Vol 9 (3) ◽

Cited By ~ 9

Author(s):

Neal P. Dillon ◽

Ramya Balachandran ◽

J. Michael Fitzpatrick ◽

Michael A. Siebold ◽

Robert F. Labadie ◽

...

Keyword(s):

Computed Tomography ◽

High Speed ◽

Degrees Of Freedom ◽

Compact Bone ◽

Target Volume ◽

Surgical Workflow ◽

The Mean ◽

Milling Robot ◽

High Level ◽

Temporal Bones

Otologic surgery often involves a mastoidectomy, which is the removal of a portion of the mastoid region of the temporal bone, to safely access the middle and inner ear. The surgery is challenging because many critical structures are embedded within the bone, making them difficult to see and requiring a high level of accuracy with the surgical dissection instrument, a high-speed drill. We propose to automate the mastoidectomy portion of the surgery using a compact, bone-attached robot. The system described in this paper is a milling robot with four degrees-of-freedom (DOF) that is fixed to the patient during surgery using a rigid positioning frame screwed into the surface of the bone. The target volume to be removed is manually identified by the surgeon pre-operatively in a computed tomography (CT) scan and converted to a milling path for the robot. The surgeon attaches the robot to the patient in the operating room and monitors the procedure. Several design considerations are discussed in the paper as well as the proposed surgical workflow. The mean targeting error of the system in free space was measured to be 0.5 mm or less at vital structures. Four mastoidectomies were then performed in cadaveric temporal bones, and the error at the edges of the target volume was measured by registering a postoperative computed tomography (CT) to the pre-operative CT. The mean error along the border of the milled cavity was 0.38 mm, and all critical anatomical structures were preserved.

Download Full-text

THE STUDENTS’ COMPREHENSION IN ANALYZING ENGLISH TEXT OF THE MA DARUL FURQAN NW MENGKURU GRADERS

JUPE : Jurnal Pendidikan Mandala ◽

10.36312/jupe.v4i5.1326 ◽

2019 ◽

Vol 4 (5) ◽

Author(s):

Mul Muliadi

Keyword(s):

English Text ◽

Average Level ◽

Low Level ◽

The Mean ◽

High Level ◽

Very High

This research is aimed to measure the students’ compression in analyzing English text for the students of MA Darul Furqan NW Mengkuru. The students’ comprehension in analyzing English text for the tenth year students of MA Darul Furqan NW Mengkuru are in average level. It can be seen from the mean score of the students that is 31.35 in which this number belongs average level. The percentages of successes of students’ comprehension in analyzing English text for the tenth year students of MA Darul Furqan NW Mengkuru are low. After the scores were classified for the students’ comprehension in analyzing English text, the researcher found 4 students who got very high score. It means that there were 13.33% of them were categorized very high level. Furthermore, there were 4 students who got high level; it means that there were 13.33% of students who were categorized high level. There were7 students who got sufficient level; it means that there were 23.33% of students who were categorized high level, and there were 15 students who got low level; it means that there were 50% of students who were categorized low level, moreover, there were none of students who got very.

Download Full-text

Cloud climatologies from the infrared sounders AIRS and IASI: strengths and applications

Atmospheric Chemistry and Physics ◽

10.5194/acp-17-13625-2017 ◽

2017 ◽

Vol 17 (22) ◽

pp. 13625-13644 ◽

Cited By ~ 8

Author(s):

Claudia J. Stubenrauch ◽

Artem G. Feofilov ◽

Sofia E. Protopapadaki ◽

Raymond Armante

Keyword(s):

Surface Temperature ◽

Cloud Amount ◽

Co2 Concentration ◽

Boreal Summer ◽

Ancillary Data ◽

Low Level ◽

Geographical Patterns ◽

Cloud Properties ◽

The Mean ◽

High Level

Abstract. Global cloud climatologies have been built from 13 years of Atmospheric Infrared Sounder (AIRS) and 8 years of Infrared Atmospheric Sounding Interferometer (IASI) observations, using an updated Clouds from Infrared Sounders (CIRS) retrieval. The CIRS software can handle any infrared (IR) sounder data. Compared to the original retrieval, it uses improved radiative transfer modelling, accounts for atmospheric spectral transmissivity changes associated with CO2 concentration and incorporates the latest ancillary data (atmospheric profiles, surface temperature and emissivities). The global cloud amount is estimated to be 0.67–0.70, for clouds with IR optical depth larger than about 0.1. The spread of 0.03 is associated with ancillary data. Cloud amount is partitioned into about 40 % high-level clouds, 40 % low-level clouds and 20 % mid-level clouds. The latter two categories are only detected in the absence of upper clouds. The A-Train active instruments, lidar and radar of the CALIPSO and CloudSat missions, provide a unique opportunity to evaluate the retrieved AIRS cloud properties. CIRS cloud height can be approximated either by the mean layer height (for optically thin clouds) or by the mean between cloud top and the height at which the cloud reaches opacity. This is valid for high-level as well as for low-level clouds identified by CIRS. IR sounders are particularly advantageous to retrieve upper-tropospheric cloud properties, with a reliable cirrus identification, day and night. These clouds are most abundant in the tropics, where high opaque clouds make up 7.5 %, thick cirrus 27.5 % and thin cirrus about 21.5 % of all clouds. The 5 % annual mean excess in high-level cloud amount in the Northern compared to the Southern Hemisphere has a pronounced seasonal cycle with a maximum of 25 % in boreal summer, in accordance with the moving of the ITCZ peak latitude, with annual mean of 4° N, to a maximum of 12° N. This suggests that this excess is mainly determined by the position of the ITCZ. Considering interannual variability, tropical cirrus are more frequent relative to all clouds when the global (or tropical) mean surface gets warmer. Changes in relative amount of tropical high opaque and thin cirrus with respect to mean surface temperature show different geographical patterns, suggesting that their response to climate change might differ.

Download Full-text

Effect of Vertical Stretching on the Extraction of Mean Identity From Faces

Perception ◽

10.1177/0301006617701160 ◽

2017 ◽

Vol 46 (9) ◽

pp. 1048-1061 ◽

Cited By ~ 1

Author(s):

Alexandra Bagaïni ◽

Graham Hole

Keyword(s):

Physical Characteristics ◽

Current Experiment ◽

Target Condition ◽

Identity Information ◽

Low Level ◽

The Face ◽

The Mean ◽

Face Stimuli ◽

High Level ◽

Trial Participants

Observers can extract the mean identity from a set of faces and falsely recognise it as a genuine set member. The current experiment demonstrated that this ‘perceptual averaging’ also occurs with vertically stretched faces. On each trial, participants decided whether a target face was present in a preceding set of four faces. In the control condition, the faces were all normally proportioned; in the stretched set condition, the face sets were stretched but the targets were normal; and in the stretched target condition, the face sets were normal but the targets were stretched. In all three conditions, participants falsely identified the set mean as a face that had been presented within the set, implying that this identity-averaging effect is based on high-level identity information rather than the low-level physical characteristics of the face stimuli.

Download Full-text

TEACHING METHODS, ACHIEVEMENT AND HIGH ORDER THINKING SKILLS (HOTS) AMONG ACCOUNTING STUDENTS IN SECONDARY SCHOOLS IN MALAYSIA

International Journal of Education, Psychology and Counseling ◽

10.35631/ijepc.4330011 ◽

2019 ◽

Vol 4 (33) ◽

pp. 132-142

Author(s):

Norlia Mat Norwani ◽

Rohaila Yusof ◽

Rusliza Yahya ◽

Zuriadah Ismail

Keyword(s):

Teaching Methods ◽

Thinking Skills ◽

Accounting Students ◽

Teaching Activities ◽

Low Level ◽

Teaching Aids ◽

Quantitative Skills ◽

The Mean ◽

High Level ◽

High Order Thinking

The purpose of this study is to determine the correlation between Accounting teachers’ use of various teaching methods with achievement and high order thinking skills (HOTS) of Principles of Accounting students. Samples are formed four students from schools in the Batang Padang district in Perak, Malaysia. The objectives of the study are to determine the level of practice of teaching methods, student achievement, and HOTS, the difference of the variables based on gender and quantitative skills and the correlation between teacher teaching methods and student achievement and HOTS. The practice of teaching methods is divided into three constructs namely teaching activities, teaching aids, and teaching and facilitating. The findings show that the mean level of teaching activities is 3.79 (high), teaching aids 4.41 (high) and teaching and facilitating 3.57 (moderate). The overall level of students' achievement is 57.26% at a moderate level. The mean of low-level thinking skills is 5.49, medium-level 32.97 and high-level 13.31. Based on gender, students’ achievement and HOTS differ significantly with male students showing higher scores. Based on the students' quantitative skills, teaching and facilitating, low-level thinking skills and high-level thinking skills show significant differences. The Pearson correlation test between teaching aids and low-level thinking skills showed a significant positive relationship (r = 0.260, p <0.05). Based on the findings, grouping activities such as cooperative learning and systematic practice modules are expected to enhance Accounting students’ mastery and HOTS. The use of game-based learning and interactive teaching methods will have a better impact on students’ learning considering their high interest in such activities. Adequate infra and info structures and tools that facilitate the use of ICT in teaching and facilitating should be made available. Teachers should also ensure variability in teaching activities and teaching aids to enhance students’ motivation and performance.

Download Full-text

The influence of temperature on the survival and infectivity of the cercariae of Echinoparyphium recurvatum (Digenea: Echinostomatidae)

Parasitology ◽

10.1017/s003118209900400x ◽

1999 ◽

Vol 118 (4) ◽

pp. 383-388 ◽

Cited By ~ 59

Author(s):

A. M. McCARTHY

Keyword(s):

Death Rate ◽

Transmission Efficiency ◽

Effect Of Temperature ◽

Survival Times ◽

Temperature Dependent ◽

Natural Habitats ◽

Low Level ◽

The Mean ◽

Per Capita ◽

High Level

Survival and infectivity characteristics are described for cercariae of the echinostome Echinoparyphium recurvatum at 10, 15, 20, 25 and 30°C. Cercarial survival was markedly age- and temperature-dependent, maximum survival time being reduced from 68 h at 10°C to 12 h at 30°C, 50% survival times falling from 47·9 h at 10°C to 8·4 h at 30°C. The mean instantaneous per capita death rate of cercariae increased exponentially with temperature. Cercarial infectivity was also temperature dependent, the main effect of temperature being upon the rate at which infectivity diminished with increasing age of cercariae. Infectivity increased from zero at 10°C to a maximum at 25°C after which it declined to a low level at 30°C. Overall transmission efficiency of cercariae at each experimental temperature was calculated as the ratio of the mean instantaneous per capita death rate (μ) to the mean instantaneous per capita rate of infection (β). Transmission efficiency of cercariae was found to be maximal at 20°C and at a relatively high level between 10 and 25°C. Transmission efficiency was zero at 10°C and at a very low level at 30°C. These results suggest that the cercariae of E. recurvatum show transmission optima at water temperatures likely to be encountered in natural habitats in Britain and Europe.

Download Full-text

WHAT YOUNG INDIA WANTS: STUDYING THE EFFECT OF LEVEL OF CSR ENGAGEMENT ON ORGANIZATIONAL ATTRACTIVENESS

10.36106/1716004 ◽

2021 ◽

pp. 24-27

Author(s):

Sriparna Gogoi ◽

Saumya Singh ◽

Abinash Kashyap

Keyword(s):

Repeated Measures ◽

Indian Population ◽

Organizational Attractiveness ◽

Repeated Measures Anova ◽

Low Level ◽

The Mean ◽

High Level ◽

Post Hoc ◽

A Company ◽

Csr Engagement

This study examined the effect of level of engagement in CSR activities on organizational attractiveness as perceived by a young Indian population. A sample of 120 participants between the ages of 18-23 years was assigned to one of three groups, each group containing 40 participants. Participants in each group were sent an online from containing a company prole and a questionnaire that assessed how attractive the participants found the company as potential place to work. The prole sent to each group differed in their descriptions of the company's CSR activities with one prole describing a high level of engagement in CSR, the second describing a low level of engagement and a third with no description of CSR activities. The mean scores on organizational attractiveness given by the three groups to their respective proles were compared using one-way repeated measures ANOVA and Tukey's HSD Post-hoc test. Findings showed a signicant difference in the mean scores of the three proles, with the High CSR company prole being signicantly higher on organizational attractiveness than the Low CSR prole. However, difference in the mean scores of the High CSR prole and the No CSR prole were not signicant. Results indicate that engagement in CSR activities inuenced respondents' evaluation of the companies and a higher amount of CSR activity was perceived favourably in conditions wherein the proles mentioned the company's CSR work

Download Full-text