Research and Application of Sun Shadow Location Technology in Video Big Data

2017 2nd International Conference on Computer Engineering, Information Science and Internet Technology (CII 2017) ISBN: 978-1-60595-504-9

Research and Application of Sun Shadow Location

Technology in Video Big Data

WEIHU WANG, ZENGGANG XIONG and YANSHEN LIU

ABSTRACT

This paper presents a sun shadow location model for quickly determining and analyzing the location and time of the photographed object in video big data. Because the length of the object's shadow usually changed, it is related to some factors, such as the solar zenith angle, the solar azimuth angle and the geographical latitude of the object. So it proposes building the location information fingerprint database, which shows mainly the relation between the length of the object’s shadow and time by reasonable assumptions for the interception of the video data. Then, it is very convenient to predict the location and time of the taken object by indexing the target in database. Finally, some experimental results show that the proposed model can accurately locate the object.

KEYWORDS

Sun shadow; Location; Video Big Data

INTRODUCTION

At present, intelligent video analysis technology has been widely used in all kinds of social domains. It mainly contains three kinds of important technology, such as target detection, video positioning and behavior recognition [1]. In the video positioning, Mean shift object tracking algorithm and Kalman filter positioning method are often used, which can achieve the geographic position (coordinates)[2]. But there are few studies about the location of the time and identity information where the object is at the moment.

So, this paper presents a sun shadow location model for quickly determining and analyzing the location and time of the photographed object in intelligent video big data. It can determine when and where the object is filmed in video by analyzing the sun shadow changing of the object in the video big data. That is to achieve the location of the time and position of the object photographed.

It is well known that the essence of the shadow is that the light is blocked by the sun, and the object is projected on the ground. When knowing the length of a given object, it is clear to find relation among the length of the object's shadow, the latitude and longitude, the local date and time, and the rotation of the earth.

_________________________________________

Weihu Wang, Hubei Engineering University & Hankou University, Xiaogan & Wuhan Hubei, China.

According to the latitude, longitude and date time in a place, it is possible to compute some parameters, such as declination and solar elevation angle. Then it can draw out the relation between the length of the object’s shadow and the time, in order to build sun shadow location model (the Location Information Fingerprint Database) through a series of reasonable assumptions. On the contrary, it is very convenient to predict the location and time of the photographed object by indexing the target object in database

The rest of the paper is organized as follows. Section II describes the establishment and solution of the sun shadow location model; Section III confirms the accuracy of the model by some simulation experiments.

SUN SHADOW LOCATION MODEL

Assumptions

In order to approach to the real scene, the paper presents many reasonable assumptions in the process of establishing the sun shadow location model, as follows.

Ignoring the effects of atmospheric scattering;

The orbit of the earth is circle and revolution speed of the earth is the same speed by default;

Ignoring the influence of the revolution; Solar light is a parallel light by default;

The earth is a sphere, and the ground is a flat by default; There is 365 days a year, 24 hours a day by default; Other celestial bodies don’t impact on the earth by default.

Solution

When knowing length of the object, there is an inevitable relation between the length changing of the object’s shadow and some such shot parameters such as the longitude and latitude, local date time, etc. These parameters of sun shadow location model will be solved one by one in Section 2.1 now.

Thinking

Thinking of the sun shadow location model, as shown in Figure 1.

Figure 1. Thinking of sun shadow location model.

The solution of parameters

(1)The solution of declination 

After getting shooting date, it is necessary to convert into the number of days in a year, recorded as n (based on January 1st in the year), so it can get the number of arc occupied by this day, as shown expression (1).





2 1

365

pi n

b  

(1)

Note:

: Circumference Ratio .The value is 3.1415926; :The number of days in a year;

:A day occupied by the number of the arc.

According to the formula of the direct point of the sun, as shown in the formula (2).

 

 

 

 

 

 





0.006918 0.399912cos 0.070257sin

0.006758cos 2 0.000907sin 2 180 /

0.002697cos 3 0.00148sin 3

b b

b b pi

b b



 

 

 

 _  _

_{ } 

 

(2)

Note:

 Declination. It is the angle between the earth's equatorial plane and the connection of the sun and the center of the earth.

It is able to find out the sun declination of the shooting date by formula (2).

(2) The solution of the solar height angle hs

In the analysis of the length changing of the object′s shadow, it can be found that the longitude of the object has not changed, but the latitude has changed. So this paper takes the sectional view, as shown in figure 2.

Figure 2. The sectional view.

Figure 3. The relationship among the three parameters.

When sunlight shines on an object, which has H high on the ground, the sun and the object’s shadow forms an angle, which is less than hs. With the change of the sun’

position, the length of the object’s shadow also changes, so the angle hs is constantly changing as well.

Note:

hs_{: The solar altitude angle.} _{: The hour angle.}

_{: The angle between the sun and the equator;}

_{: Declination. The angle between the line of the object and the center of the}

earth and the equator.

Based on the formula (2) and formula (3), it is convenient to obtain the solar height angle of the object at a position.

 

sin hs     ₍₃₎

(3) The solution of shadow’s length

According to Fig 2, the position relation of the sun, the object, and the object’s shadow, is shown in Figure 3.

It can be obtained from Fig 3, as shown.

tanhs H L



tan

H L

hs

Note:

H: The object's height; L: The length of the object’s shadow

(4) The solution of Sun Shadow Location Model

To sum up, the geometric math model of the Sun Shadow Location is as follows:

SIMULATION EXPERIMENTS

Simulated data

In the studying process, it is possible to establish the Sun Shadow Location Model through measuring the length of the bar’s shadow in Beijing.

The shooting place is Tian’anmen square, which is located at latitude39 54 26  , longitude116 23 29   .The shooting date is from 9:00 to 15:00 in October 22, 2015.The length of bar is 3m.

In the experiment, it can transfer a video to some serial frames of image, which are stored by MATLAB software. It had already acquired some data about the shadows of objects with the time variation, what I call “Location Fingerprint Information Database”, as shown in Table 1.

The solution of Sun Shadow Location Model

It is concluded that the shadow of the object along with the time change curve by MATLAB software and some related parameters, as shown in Figure 4.

TABLE 1. LOCATION FINGERPRINT INFORMATION DATABASE.

Notes:

The horizontal coordinate represents time;

TABLE 2. LOCATION DATA OF TIME AND POSITION (KNOWN OBJECT'S SHADOW).

Note:

The south of latitude and longitude is the negative, the north is positive.

The longitudinal axis represents the length of the bar’ shadow;

In Figure 4, it can be clearly found that the length of the bar’s shadow is various from 9:00 to 12:00; it is the shortest in the 12:00; then the length of the bar’s shadow is slowly growing from 12:00 to 15:00.

Location Model

According to the Sun Shadow Location Model, it is convenient to compute the latitude, longitude and time of the object located, so that it is possible to accomplish positioning of the object.

In the simulation environment, the object is still located Tian’anmen square in Beijing, which is located at latitude 39 54 26  , longitude 116 23 29   (longitude remains unchanged). Given the length of the object’ shadow in 9:00-15:00 October 1, 2015, so that it can determine shooting time and location of objects. The experimental data such as table 2 shows.

CONCLUSION

This paper presents a sun shadow location model for quickly determining and analyzing the location and time of the photographed object in intelligent video big data analysis. Combining with the length changing of the object's shadow is related to some factors, It proposes building the l Location Fingerprint Information Database (LFID). It is very convenient to predict the location and time of the taken object by indexing the target in database. Finally, some experimental results also show that the proposed model can accurately locate the object.

ACKNOWLEDGMENTS

Correspondence author: Zenggang Xiong.

Fund Name: Youth innovation team project in Hubei Provincial Department of Education. Fund No: T201410.

REFERENCES

1. Li Zhongjie; Li Siwei; Yin Rui-rui; Jiang yu; Journal of Chengdu Normal University,2016(05) 2. Xu Jianyuan. Research and application of the key technology in intelligent video analysis[D].

University of Electronic Science and Technology of China, 2013.

4. Jiang Hongli. Mathematical deduction and analysis of the solar direct point latitude[J]. Mathematical Bulletin, 2007(9):39-40.

5. Wang Guoan, Mi Hongtao. Calculation of the annual variation range of the solar azimuth and the sun azimuth angle at sunrise and sunset time[J]. Meteorology and Environmental Science, 2007, 30(B09):161-164.

6. Chen Jianting, Wen Yinting, Fu Shouzhong. Calculation of optimum solar azimuth[J]. Journal of Zhaoqing University, 2014, 35(2):20-22.

7. Li Lin, Hu Yong, Gong Cai Lan, Ling Ya Zhu, He Huaying. Influence of solar altitude angle on image energy and its correction[J]. Journal of atmospheric and Environmental Optics, 2013, 8(1):11-17.

8. Wang Bingzhong, Tang Jie. Comparison of several methods for calculating the position of the sun[J]. Journal of Solar Energy, 2001, 22:413-417.

9. SHEN Yi-fan ZHAO Feng LI Chen-hui. Research on the Location Model of Solar Shadow Based onMATLAB[J]. Value Engineering, 2016, 35(18):74-76.

10. Huang Xiaohui, Zhu Jiaming, Weng Jin, Fang Tingting. The research of positioning technology of the model of sun's shadow based on the view of space-time[J]. Journal of Qiqihar University (Natural Science Edition), 2016(1):68-72.

11. .Zhu Haiyang,Zhang Mengqi. Application of sun shadow location algorithm in video data analysis[J]. Information & Communications, 2015(12):29-30.