scholarly journals Correction to: A Deep-Fusion Network for Crowd Counting in High-Density Crowded Scenes

2021 ◽  
Vol 14 (1) ◽  
Author(s):  
Sultan Daud Khan ◽  
Yasir Salih ◽  
Basim Zafar ◽  
Abdulfattah Noorwali
Keyword(s):  
High Density ◽  
Crowd Counting ◽  
Crowded Scenes

scholarly journals A High-robustness and Low Resource-consumption Crowd Counting Model

2021 ◽  
Vol 15 ◽  
pp. 46-54
Author(s):  
Han Jia ◽  
Xuecheng Zou
Keyword(s):  
Network Architecture ◽  
Resource Consumption ◽  
Portable Devices ◽  
Estimation Errors ◽  
Crowd Counting ◽  
Density Map ◽  
Almost All ◽  
Fully Connected ◽  
Crowded Scenes ◽  
Counting Model

A major problem of counting high-density crowded scenes is the lack of flexibility and robustness exhibited by existing methods, and almost all recent state-of-the-art methods only show good performance in estimation errors and density map quality for select datasets. The biggest challenge faced by these methods is the analysis of similar features between the crowd and background, as well as overlaps between individuals. Hence, we propose a light and easy-to-train network for congestion cognition based on dilated convolution, which can exponentially enlarge the receptive field, preserve original resolution, and generate a high-quality density map. With the dilated convolutional layers, the counting accuracy can be enhanced as the feature map keeps its original resolution. By removing fully-connected layers, the network architecture becomes more concise, thereby reducing resource consumption significantly. The flexibility and robustness improvements of the proposed network compared to previous methods were validated using the variance of data size and different overlap levels of existing open source datasets. Experimental results showed that the proposed network is suitable for transfer learning on different datasets and enhances crowd counting in highly congested scenes. Therefore, the network is expected to have broader applications, for example in Internet of Things and portable devices.

scholarly journals A High-Density Crowd Counting Method Based on Convolutional Feature Fusion

2018 ◽  
Vol 8 (12) ◽  
pp. 2367 ◽  
Author(s):  
Hongling Luo ◽  
Jun Sang ◽  
Weiqun Wu ◽  
Hong Xiang ◽  
Zhili Xiang ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Large Scale ◽  
Feature Fusion ◽  
High Density ◽  
Counting Problem ◽  
Crowd Counting ◽  
Low Level ◽  
Density Map ◽  
High Level

In recent years, the trampling events due to overcrowding have occurred frequently, which leads to the demand for crowd counting under a high-density environment. At present, there are few studies on monitoring crowds in a large-scale crowded environment, while there exists technology drawbacks and a lack of mature systems. Aiming to solve the crowd counting problem with high-density under complex environments, a feature fusion-based deep convolutional neural network method FF-CNN (Feature Fusion of Convolutional Neural Network) was proposed in this paper. The proposed FF-CNN mapped the crowd image to its crowd density map, and then obtained the head count by integration. The geometry adaptive kernels were adopted to generate high-quality density maps which were used as ground truths for network training. The deconvolution technique was used to achieve the fusion of high-level and low-level features to get richer features, and two loss functions, i.e., density map loss and absolute count loss, were used for joint optimization. In order to increase the sample diversity, the original images were cropped with a random cropping method for each iteration. The experimental results of FF-CNN on the ShanghaiTech public dataset showed that the fusion of low-level and high-level features can extract richer features to improve the precision of density map estimation, and further improve the accuracy of crowd counting.

scholarly journals Deep Learning-Based Crowd Scene Analysis Survey

2020 ◽  
Vol 6 (9) ◽  
pp. 95
Author(s):  
Sherif Elbishlawi ◽  
Mohamed H. Abdelpakey ◽  
Agwad Eltantawy ◽  
Mohamed S. Shehata ◽  
Mostafa M. Mohamed
Keyword(s):  
Deep Learning ◽  
Automatic System ◽  
Scene Analysis ◽  
Crowd Counting ◽  
Public Events ◽  
Analysis Methods ◽  
The Difference ◽  
Evaluation Metric ◽  
Crowded Scenes

Recently, our world witnessed major events that attracted a lot of attention towards the importance of automatic crowd scene analysis. For example, the COVID-19 breakout and public events require an automatic system to manage, count, secure, and track a crowd that shares the same area. However, analyzing crowd scenes is very challenging due to heavy occlusion, complex behaviors, and posture changes. This paper surveys deep learning-based methods for analyzing crowded scenes. The reviewed methods are categorized as (1) crowd counting and (2) crowd actions recognition. Moreover, crowd scene datasets are surveyed. In additional to the above surveys, this paper proposes an evaluation metric for crowd scene analysis methods. This metric estimates the difference between calculated crowed count and actual count in crowd scene videos.

scholarly journals Crowd Counting in Low-Resolution Crowded Scenes Using Region-Based Deep Convolutional Neural Networks

IEEE Access ◽  
2019 ◽  
Vol 7 ◽  
pp. 35317-35329 ◽  
Author(s):  
Muhammad Saqib ◽  
Sultan Daud Khan ◽  
Nabin Sharma ◽  
Michael Blumenstein
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Low Resolution ◽  
Deep Convolutional Neural Networks ◽  
Crowd Counting ◽  
Crowded Scenes

scholarly journals Crowd Counting Guided by Attention Network

Information ◽  
2020 ◽  
Vol 11 (12) ◽  
pp. 567
Author(s):  
Pei Nie ◽  
Cien Fan ◽  
Lian Zou ◽  
Liqiong Chen ◽  
Xiaopeng Li
Keyword(s):  
Spatial Distribution ◽  
State Of The Art ◽  
Attention Mechanism ◽  
Attention Network ◽  
Crowd Counting ◽  
Feature Map ◽  
Density Maps ◽  
Feature Extractor ◽  
Global And Local ◽  
Crowded Scenes

Crowd Crowd counting is not simply a matter of counting the numbers of people, but also requires that one obtains people’s spatial distribution in a picture. It is still a challenging task for crowded scenes, occlusion, and scale variation. This paper proposes a global and local attention network (GLANet) for efficient crowd counting, which applies an attention mechanism to enhance the features. Firstly, the feature extractor module (FEM) uses the pertained VGG-16 to parse out a simple feature map. Secondly, the global and local attention module (GLAM) effectively captures the local and global attention information to enhance features. Thirdly, the feature fusing module (FFM) applies a series of convolutions to fuse various features, and generate density maps. Finally, we conduct some experiments on a mainstream dataset and compare them with state-of-the-art methods’ performances.

Planar defects in ordered (Ga,In)P

1988 ◽  
Vol 46 ◽  
pp. 902-903
Author(s):  
S. McKernan ◽  
C. B. Carter ◽  
D. Bour ◽  
J. R. Shealy
Keyword(s):  
Electron Diffraction ◽  
Vapor Phase ◽  
Growth Direction ◽  
High Density ◽  
Ordered Structure ◽  
Planar Defects ◽  
Diffraction Patterns ◽  
Ternary Semiconductors ◽  
Anion Species ◽  
Metallic Vapor

The growth of ternary III-V semiconductors by organo-metallic vapor phase epitaxy (OMVPE) is widely practiced. It has been generally assumed that the resulting structure is the same as that of the corresponding binary semiconductors, but with the two different cation or anion species randomly distributed on their appropriate sublattice sites. Recently several different ternary semiconductors including AlxGa1-xAs, Gaxln-1-xAs and Gaxln1-xP1-6 have been observed in ordered states. A common feature of these ordered compounds is that they contain a relatively high density of defects. This is evident in electron diffraction patterns from these materials where streaks, which are typically parallel to the growth direction, are associated with the extra reflections arising from the ordering. However, where the (Ga,ln)P epilayer is reasonably well ordered the streaking is extremely faint, and the intensity of the ordered spot at 1/2(111) is much greater than that at 1/2(111). In these cases it is possible to image relatively clearly many of the defects found in the ordered structure.

Quantitative defect studies in semiconductor TEM samples prepared by low angle ion milling

1995 ◽  
Vol 53 ◽  
pp. 512-513
Author(s):  
L. Mulestagno ◽  
J.C. Holzer ◽  
P. Fraundorf
Keyword(s):  
Sample Preparation ◽  
Milling Time ◽  
High Density ◽  
Low Density ◽  
Ion Milling ◽  
High Quality ◽  
Variable Angle ◽  
Major Disadvantage ◽  
Induced Defects

Due to the wealth of information, both analytical and structural that can be obtained from it TEM always has been a favorite tool for the analysis of process-induced defects in semiconductor wafers. The only major disadvantage has always been, that the volume under study in the TEM is relatively small, making it difficult to locate low density defects, and sample preparation is a somewhat lengthy procedure. This problem has been somewhat alleviated by the availability of efficient low angle milling.Using a PIPS® variable angle ion -mill, manufactured by Gatan, we have been consistently obtaining planar specimens with a high quality thin area in excess of 5 × 104 μm2 in about half an hour (milling time), which has made it possible to locate defects at lower densities, or, for defects of relatively high density, obtain information which is statistically more significant (table 1).

Using the scanning electron microscope for failure analysis of high density magnetic media

1987 ◽  
Vol 45 ◽  
pp. 392-393
Author(s):  
Evelyn R. Ackerman ◽  
Gary D. Burnett
Keyword(s):  
Electron Microscope ◽  
Scanning Electron Microscope ◽  
Failure Analysis ◽  
High Density ◽  
Magnetic Media ◽  
Specific Data ◽  
Retrieval Systems ◽  
Operating Environments ◽  
The Media ◽  
Scanning Electron

Advancements in state of the art high density Head/Disk retrieval systems has increased the demand for sophisticated failure analysis methods. From 1968 to 1974 the emphasis was on the number of tracks per inch. (TPI) ranging from 100 to 400 as summarized in Table 1. This emphasis shifted with the increase in densities to include the number of bits per inch (BPI). A bit is formed by magnetizing the Fe203 particles of the media in one direction and allowing magnetic heads to recognize specific data patterns. From 1977 to 1986 the tracks per inch increased from 470 to 1400 corresponding to an increase from 6300 to 10,800 bits per inch respectively. Due to the reduction in the bit and track sizes, build and operating environments of systems have become critical factors in media reliability.Using the Ferrofluid pattern developing technique, the scanning electron microscope can be a valuable diagnostic tool in the examination of failure sites on disks.

Sign in / Sign up

Export Citation Format

Share Document