Open Access
Add to BibSonomy
Abstract
For the safety, underground facilities are required to be inspected regularly, especially with image analysis. Traditional wireless and wired transmission techniques have a weakness of limited transmission range in narrow underground environments. In this study, a new image transmission method based on visible light communication (VLC) has been thus proposed. Two types of detectors as an image signal receiver have been tested and discussed in the following experiments. The photodiodes (PDs) are widely used as a common image signal detector in VLC technology, but image signal detection using solar panels (SPs) has not been studied. PDs have a higher sensitivity and faster response time but a limited detection area and high cost. Besides, PDs require the lens to focus light. On the other hand, SPs have much larger optical signal receiving areas and stronger optical signal capture capabilities. They can realize lens-free detection and are inexpensive. These features of PD were firstly verified in experiments with several receiving areas and angles of detectors. The experimental result revealed that PD had better image detection and recovery capabilities than those of SP. Then, we found that a larger receiving area obtained by using double PDs/SPs improved the brightness of the restored image. In a supplementary experiment, the influence of different RGB optical components on VLC, especially the VLC-based image transmission, has been investigated by using two-dimensional Fourier transform frequency analysis. We found that the red optical component significantly increased the intensity and energy of the restored image as the image low-frequency signals were larger than the restored image using ordinary mixed white light, and moreover, the blue optical component decreased the low-frequency part of the image.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
The underground facilities are usually designed for storing and transporting resources such as oil, water, mineral, gas, electricity, etc. Once the damage brought by natural and human factors happen, it will inevitably cause waste of resources and even environmental pollution. Therefore, the regular image analysis and inspection within facilities is required. Due to the complex environment, image transmission technology has been regarded as a huge problem. Some problems exist with traditional transmission methods. For wired transmission methods, the weight and friction of cable can restrict the transmission distance especially in the narrow and tortuous underground environment [1,2]. For other normal wireless methods based on radio frequency (RF) signal, some minor problems such as unstable signal, low transmission rate, and limited transmission range still happen [3,4]. Therefore, in this study, a new image transmission approach based on visible light communication (VLC) has been proposed to improve image transmission performance. In complex underground environments, the VLC method owns several potential advantages containing higher transmission security, higher transmission rate, lower attenuation rate in the medium (e.g., water, petroleum, and dust), and stronger anti-electromagnetic interference from power-transfer cables. Especially, it can provide an illuminance for inspection, hence improve the quality of captured images [5–7].
In the previous study, positive–intrinsic–negative (PIN) diodes or avalanche photodiodes (APDs) are the commonly used photodiodes (PDs) for receiving the image signal. It was usually adopted in the stationary point-to-point transmission due to the limited receiving or detection area. Because of the low optical reception efficiency, in the transmission experiments, the convex lenses are usually used to focus the light, which requires accurate pointing between the transmitter and receiver. In [8–13], the normal PDs have been used to detect the wireless optical signal in some complex environments, such as underwater, in-home, etc. Although the communication rate could reach almost Gbit/s level by adopting Orthogonal Frequency Division Multiplexing (OFDM) or Pulse Width Modulation (PWM) methods, the stationary point-to-point experimental setup would restrict the wide application communication of VLC systems. In practical underground environments, the mobility of the image transmission system could make the point-to-point alignment more difficult and furthermore affect the system performance [14]. Recently, solar panels (SPs) have been used as an optical signal detector for VLC since they own lower price, much larger optical signal receiving areas, and stronger optical signal capture capabilities. They can also realize more flexible lens-free point-to-point signal receiving in the VLC system. The most important is that solar panel can realize signal detection and energy harvesting simultaneously in complex environments with difficulty in power supply. This function can finally improve energy efficiency. In [15–17], the works illustrate the possibility of SP-based optical communication and self-powering with precise electrical model analysis. Moreover, in [18], the potential of SP receiver with the advantages of large signal detection area and lens-free signal receiving function has been further investigated in the practical complex environments such as in the water and dust. Besides, the problem of point-to-point alignment has been improved.
The large-area SPs can somehow solve alignment problems. The most important is that SPs can play a role of both detecting optical signals for VLC and energy harvesting at the same time. This is the original motivation we have adopted SPs as receivers for our study. As the first step of this study, we focus on newly developing a VLC-based image transmission system. In the future, we will develop a system integrating image transmission with energy transmission based on VLC and solar panel. In several previous studies, the solar panel has been only used to detect the ordinary signal. In this study, the image receiving and detection with solar panel are initially considered and investigated. Through the experiments, we considered two factors including different receiving area and receiving angle to test both photodiode (PD) and solar panel (SP). In order to evaluate the PD-based or SP-based image receiving performance quantitatively, the image quality evaluation framework was built based on the comparison between the final received images and original images. This framework focuses on evaluating image transmission performance through the image quality assessment parameters such as mean square error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM). The MSE is calculated as follows:
Besides, the calculation equations of SSIM are given as follows:
2. Experimental setup
Figure 1 illustrates the experimental setup of the proposed image transmission system using photodiode and solar panel as the detectors. In the transmitter, a National Television System Committee (NTSC) format camera which owned 2.8 mm lens and a 170-degree wide angle was used for the image information collection. The analog image signals captured by the camera were added into the LED spotlight driver device. The driver device had the function of amplifying the analog signal to drive the LED array. The LED could affect the transmission distance and quality of the VLC system. It is the special XLamp XHP-70 LED array with maximum 3930 lm output which was suitable for the light source for our following experiments. In the receiver, two kinds of receivers were selected to collect the optical signal. Among them, the photodiode used Si-Pin fixed gain detector PDA10A. It was a special detector with the wavelength response range from 200 nm to 1100 nm. The active response area is about 0.8 $\textrm{m}{\textrm{m}^2}$. The signal bandwidth could reach about 150 MHz and the maximum photoelectric conversion efficiency could even reach about 0.44 A/W. To concentrate light and avoid the divergence of light, a convex lens was employed with a diameter of 63 mm and a focal length of 150 mm. Another solar panel detector was used in detecting the image optical signal. It owned maximum power of 2 W. When reaching the maximum power, the output voltage and short-circuit current could reach 5.9 V and 0.39 A, respectively. Since solar panel owns large receiving area, it did not have the requirement of point-to-point alignment, therefore it could receive the optical signal without lens [20–23]. Through the photoelectric conversion by the detectors, a key-press variable electrical attenuator (ATT) KPATT2.5-90/1S-2N was used to adjust the signal amplitude. Finally, the signal could be read by the computer through an image capture card. As depicted in Figs. 2(a) and (e), a standard target image was chosen for the experiment. There were white and black rectangle strips in the middle of the image. From long to short, each color had seven rectangle strips. The SSIM of the image was calculated based on luminance, contrast, and structure. Among them, the calculation of structure could be more accurate if the regular strips were chosen. The more regular the target object, the higher the accuracy of the structure-based calculation and the smaller the calculation error of SSIM. The reason to use different size of the strip was that we wanted to observe the two kinds of factors including receiving area and angle to affect the recovery of targets with different sizes.
Fig. 1. Experimental setup of proposed VLC-based image transmission system using (a) Double SPs detector and (b) Double PDs detector.
3. Experimental results
The ideal underground facility is a special environment without any ambient light. Ambient light will affect the quality of VLC communication. In our experiment, images are transmitted through the analog optical signals, which are easy to be interfered and distorted under the influence of ambient light. Therefore, in the test, other light sources should be turned off or blocked.
Moreover, to investigate the image signal receiving performance of solar panel detector, experiments with different receiving area, angle, and distance should be performed. Among them, the transmission distance is a factor which can affect the image transmission performance. In actual former measurements, we found that the effect of image reception was good within 1.5 m. If the range was beyond 1.5 m, the receiver could not restore the image signals, and therefore it has no meaning to evaluate the effect of distance on transmission quality with parameters MSE, PSNR, and SSIM. Within 1.5 m, the transmission distance has little impact on the restoring of image signal and image quality. Therefore, we fixed the transmission distance to 1.5 m and only investigated the test with different receiving area and angle.
3.1 Test with different receiving areas and angles
We first conducted the detection ability of PD and SP with different receiving angle in an indoor environment. In each measurement, the camera was placed at a distance of 50 cm from the target image, so that the field of view could cover the entire target. The distance between the optical transmitter and the optical receiver was set at 1.5 m. Firstly, we installed a single PD or SP detector on the receiving terminal. The white optical beam passed through the air channel and illuminated on the lens. Through the focus of the lens, PD could fully detect the optical signal. As mentioned before, in the experiment, the lens was not required. By changing the receiving angle of the detector in 0$^\circ $, 30$^\circ $, and 60$^\circ $ successively, the target image was wirelessly detected and restored in the computer terminal. In the test, three image frames for each angle were captured. Three parameters including MSE, PSNR, and SSIM which could evaluate the transmission efficiency, image transmission quality, and error were recorded through the experiments. Note that these three parameters were calculated with MATLAB tool based on grayscale images which were converted from RGB images.
Figures 2 and 3 demonstrate the images restored by detectors when they were set at different receiving angle. Observing from figures, we could conclude that with the increase of receiving angle, the deterioration of image quality was aggravated. This deterioration could be reflected in the noise, structure, contrast, and brightness of each restored image. By comparing the image detection performance of double detectors and a single detector from Figs. 2 and 3, we could observe that the increased receiving area enhanced the brightness of the image, whereas it could also increase the blurriness of the image. For example, the target image in Fig. 3(c) was much brighter than the image in Fig. 2(c) since the detection area increased. However, the image definition decreased by using double PDs. The black rectangular strips in Fig. 3(c) was much more blurred than those in Fig. 2(c).
As shown in Fig. 4, MSE was employed to reveal the error between the restored image and the original transmitted image. In this figure, with the increase of receiving angle of detector, the signal noise and image distortion became much worse. Thus, MSE rose from 2046.1 to 6828.1 by using PD, while increasing from 1914.8 to 9216.6 by using SP. It could be also obtained from the figure that due to the increase of the detection angle, the effective detection area was reduced. Especially from 30$^\circ $ to 60$^\circ $, by using PD or SP, MSE increased sharply. This phenomenon could be also revealed from the restored image from Figs. 2(c) and (d) to Figs. 3(g) and (h). The restored image by using SP illustrated much higher MSE than using PD, especially in 60°. MSE of using SP was 9216.6, whereas the MSE of using PD was 6828.1. The reason is that SP owned large optical signal receiving area and stronger light capture capabilities but there was rising time delay. This delay could significantly affect signal receiving.
Besides, the noise of the image could serve as another important reference standard to evaluate the image quality. We selected PSNR since the larger the PSNR value, the lower the signal loss rate during the optical signal transmission, and finally the higher the image quality. In Fig. 5, as the receiving angle of SP/PD receiver increased, the PSNR reduced because the efficient receiving area reduced. Less than 30$^\circ $, PSNR could reach 14.126 and 13.979 by using PD and SP, respectively. However, PSNR dropped significantly to 4.781 in 60$^\circ $ with using PD. It demonstrated that using SP had higher PSNR(=6.147) than using PD because SP owned stronger light capture capabilities, which significantly affect received signal power over noise. However, the noise could also affect the recognition of target strip. In Figs. 2(d) and (h), the recognition of the rectangular strip was almost impossible when the PSNR reached less than 6.5 whether by using SP or PD.
Finally, we selected SSIM to judge the structure deterioration of restored image by detector. Generally, the higher the value of SSIM, the smaller the distortion of the received image in structure, brightness, and contrast, as well as the closer it is to the original transmitted image. Figure 2 depicted that the structure and contrast by using SP were worse than using PD. It could be also reflected in Fig. 6. In this figure, the restored image by using SP illustrated much smaller SSIM than using PD. Especially, when the angle increased to 60°, the SSIM of using SP dropped to 0.22183, which was 0.14 smaller than using PD. Although SP owned large optical signal receiving area and stronger light capture capabilities, the response rising time was much longer than PD, which affected the restored image structure because the high-frequency component of optical signal had problems. In this 60°, by using SP detector, the structure deterioration occurred, and the image quality degraded significantly making it impossible to judge from the detailed structure and brightness of the target image [24–28].
In the experiment of receiving area, a basic conclusion was obtained that since scattering and absorption happened during the light transmission in the medium, thus if we increased the size or detection area of the detectors in the experiment, the received light power and image capture capability could enhance. This could be reflected in the brightness of the images between Figs. 3(b)–(d) and (f)–(h). The images in Figs. 3(f–h) were much brighter than images in Figs. 3(b)–(d) since double SPs or PDs were used.
For comparison between double PDs/SPs and single PD/SP performance, the restored image by using single PD/SP showed higher MSE than using double PDs/SPs. Double PDs/SPs owned much stronger visible optical signal capture ability than single PD/SP because the receiving area became larger. At the same time, due to the fusion of detectors’ signal, the loss of high frequency signal will be less. For comparison between double PDs/SPs and single PD/SP performance, double PDs/SPs had larger PSNR than single PD/SP. Since they owned much larger optical signal receiving area, higher sensitivity and stronger light capture capabilities than single PD/SP. Besides, double PDs/SPs had larger SSIM than single PD/SP. Since it owned larger optical signal receiving area, higher sensitivity, and stronger light capture capabilities than single PD/SP, the brightness and contrast of each image have been improved by using double PDs/SPs.
3.2 Test with an enhanced optical component
Usually LED white light was used as the light source in the VLC technology. As a mixed light, white light can be expressed by mixing three primary light color components including red, green, and blue, and these three components have three different wavelengths and frequencies. The influence of different components on VLC especially the VLC-based image transmission has not been studied. In this study, we have adopted three wavelength bands to analyze the influence on VLC in detail. The wavelength band were red (622–770 nm), green (492–577 nm), and blue (455–492 nm), respectively. As shown in Fig. 7, two optical transmitters were placed in the test field. One was still the white mixed optical transmitter, and the other is the optical transmitter with a single component of red, green, or blue after being filtered out by the light filter. Through experiments, we investigated the effect of the enhanced optical components (such as red frequency) on the VLC-based image transmission characteristics (such as image structure, spectrum, and contrast). In the receiver, double PDs or SPs were employed to detect the optical signals from the transmitter.
Fig. 7. Experimental setup of proposed VLC-based image transmission system using enhanced R/G/B optical components.
As depicted in Fig. 8, the leftmost column corresponded to the image restored by detecting mixed white light, and the right parts of this figure illustrated the image restored by detecting mixed white with enhanced R/G/B optical components. For both using double PDs and using SPs detectors, three image frames were captured and chosen as test samples. In Fig. 8, the spectrum analysis results were also obtained for each image. This spectrum analysis was based on image Two Dimensional-Fourier Transform (2D-FT) calculation. The 2D-FT could be realized by the calculation of the following two formulas
Fig. 8. VLC-based image transmission system using enhanced R/G/B optical component and spectrogram. (a–d) are restored images received by double PDs detector, (e–h) are corresponding spectrogram. (i–l) are restored images received by double SPs detector, (m–p) are corresponding spectrogram. U-axis and V-axis of spectrogram (center part) indicates the low-frequency part of the image and area away from the both axes of the spectrogram (surrounding part) indicate the high-frequency part. The high-frequency parts can determine the outlines and details of the restored image, and the low-frequency parts can determine the intensity and brightness of the restored image.
Among the formulas, u and v represent the spatial frequencies in spectrogram, respectively. x and y are the horizontal and vertical values of the original image pixels, $f({x,y} )$ is a 2D spatial grayscale function of image defined over an X-Y plane, and $F({u,v} )$ denotes the 2D spectrum result of $f({x,y} )$. (8) is the inverse transform of (7). In (8), the complex exponentials ${e^{2j\pi ({ux + vy} )}}$ can be represented as
This formula represents a planar sinusoid in the X-Y plane of original image along the vector direction $\theta = ta{n^{ - 1}}({v/u} )$ with frequency $\omega = \sqrt {{u^2} + {v^2}} $ [29,30].
Normally, in the middle region of Fig. 8(b), the white and black rectangular strips were located in sequence on the horizontal axis. Moreover, the gray value of the image in the X-axis changed repeatedly. Assuming that if the original spatial image is a striped pattern image with regular and equal frequency changes, then, in the spectrogram, it will display two symmetrical white bright spots which can be described as the frequency value ${e^{j\pi ({ux + vy} )}}$ and ${e^{ - j\pi ({ux + vy} )}}$, respectively. Actually, since the original image is affected by irregular changes of strip pattern, ambient light, and noise, symmetrical white spots will be discretely distributed on the U-V axes of the spectrogram. Usually, the higher the frequency of gray changes in the original spatial image, the farther the two symmetrical bright spots are distributed in the spectrogram. From Fig. 8(b), we can observe that the frequency change of gray value in the X-axis is smaller than that of the Y-axis, therefore, the symmetrical frequency spots are more concentrated on the U-axis than on the V-axis in the spectrogram of Fig. 8(f). Also, according to the analysis of frequency equation $\omega = \sqrt {{u^2} + {v^2}} $, the U-axis and V-axis of spectrogram (center part) represented the low-frequency part of the image and area away from the both axes (surrounding part) of the spectrogram indicated the high-frequency part. They were marked with green and red circles in Figs. 8(e) and (m) respectively. Especially, in the center of spectrogram is the zero frequency. Moreover, the high-frequency part determined the outlines and details of the image, and the low-frequency part determined the intensity and brightness of the image.
According to the spectrogram of Figs. 8(e) and (f), by using double PDs, it was obvious that if the red optical component was enhanced, the low-frequency signal of the image in Fig. 8(f) would become larger than that low-frequency signal of the image using mixed white light in Fig. 8(e), and the intensity of image would become higher than the image using mixed white light by compared between Figs. 8(a) and (b). From Figs. 8(f) to (h), compared with blue or green optical components, the red optical component could also significantly increase the intensity and energy of the image in Fig. 8(b) since the low-frequency signal was higher than the image with enhanced blue or green optical components. This intensity change could also be reflected from Figs. 8(b) to (d). As depicted in Fig. 8(g), the green optical components could maintain the similar spectrum characteristics compared with the spectrogram of restored image using mixed white light in Fig. 8(e). The result demonstrated that the green light component has little effect on the signal frequency of the restored image. Moreover, the blue optical component would decrease the low-frequency part of the image as shown in Fig. 8(h). Although the intensity of the image was reduced, the outlines and edges were still preserved by observing from Fig. 8(d) since the high-frequency parts area in Fig. 8(h) was enhanced than Figs. 8(f) and (g).
As shown in Figs. 8(f) and (n), the low-frequency signals in the image detected by the double SPs is much larger than those in the image detected by double PDs. These low-frequency signals would make the image much brighter than image received with double PDs as shown from Figs. 8(b) and (j). Also, in similar to double PDs, the intensity and brightness of image would become higher from Figs. 8(j) and (i) with the increasing of low-frequency signal of the image if using the enhanced red optical component. Similar with the results by using PDs detector, the green optical components had little impact on spectrum characteristics as shown in Fig. 8(o). Besides, compared with red or green optical components, the blue optical component could reduce the low-frequency signal in Fig. 8(p), whereas the outlines and edges in Fig. 8(l) could be still preserved due to the increasing of high-frequency parts in Fig. 8(p) than Figs. 8(n) and (o).
3.3 Findings from experiments and discussion
In this study, it was the first attempt to evaluate the solar panel (SP) as the detector for VLC-based image transmission. SP owns a much larger optical signal receiving area and stronger light capture capabilities than photodiode (PD), which can be obviously reflected from PSNR (1.366 higher than PD in 60$^\circ $ and 2.36 higher than PD in 30$^\circ $) from Fig. 5. These outstanding performances enable lens-free image detection with higher mobility for the complex underground facilities.
Nevertheless, SP illustrates weaker detection sensitivity than PD. As observed in Fig. 2, the restored images by SP had larger greater deterioration in image structure, noise, and contrast. This deterioration can be reflected in the MSE (2388.5 higher than PD in 60$^\circ $ and 471.44 higher than PD in 30$^\circ $) and SSIM (0.14048 lower than PD in 60$^\circ $ and 0.03247 lower than PD in 30$^\circ $) from Figs. 4 and 6. In the future, the advanced SP with more sensitive panel materials should be developed and tested.
Double SPs own stronger optical signal capture capability than single-SP due to the larger signal detection area. In Fig. 5, PSNR of SPs was 4.744 larger than SP in 60$^\circ $. This means that the received signal power can be enhanced by increasing the detection area. Moreover, the brightness and contrast of each restored image were well improved as the SSIM of SPs was 0.295 higher than SP in 60$^\circ $. However, the structure of image by using SPs became worse than using a single SP. Maintaining less distortion while increasing the brightness will be the research topic of future works. A special panel with a larger receiving area and shorter signal response time should be designed.
As shown in Fig. 8, although the brightness of image increases, the sharpness and structure of the image became worse by using enhanced red optical component. However, the intensity of the image was improved by using the enhanced red optical component. To obtain some accurate image information within underground facilities including scars, breakages, cracks and other subtle features, high-brightness and high-definition images are necessary. Maintaining both high brightness and definition of the image is also an important topic which should be further studied. In the future, we will apply SP-based image transmission technology to the robot chain system which we have previously studied for underground facilities detection and maintenance [31,32].
4. Contributions and limitations
In this section, we clarify the contributions and limitations of the study.
4.1 Contributions
The contributions of the study can be summarized as mainly three topics as follows. (1) First application of VLC using solar panels (SPs) to image transmission. The SPs are firstly applied in the VLC-based image transmission since they own larger optical signal sensing receiving area and relatively cheaper compared with photodiodes (PDs). At present, there are very few studies on SPs for detecting optical signals such as control commands or sensor data, also few studies on PDs for receiving image optical signals. Thus, the investigation on SPs for VLC-based image signal detection is completely new. The ultimate target of our study is to use SPs as signal receivers for image transmission in special environments such as underground facilities, ocean, underground mine, etc. in the future. Our preliminary investigation was imperfect, but it could be a great step. (2) Quantitative and comparative evaluation. Then, the image signal receiving performance of SP detector have been quantitatively evaluated through various kinds of comparative experiments, by using standard image evaluation parameters including MSE, PSNR, and SSIM. As preliminary contributions, we found that SP owns large optical signal receiving area and strong light capture capabilities which significantly affect received signal power over noise, but it has a rising time delay which significantly affect image signal restoring. (3) Investigation on the influence of optical components. The influence of different optical RGB components on VLC, especially the VLC-based image transmission, has been firstly studied. Usually in VLC, we adopt white visible light, but it is not clear which components extracted from white light affect the performance of VLC-based image transmission. Through our study, we could obtain that the red optical component significantly increases the intensity and energy of the restored image. Moreover, the blue optical component decreases the low-frequency part of the image for the system. These new findings will support us to improve the efficiency of VLC-based image transmission in the future.
4.2 Limitations
On the other hand, this preliminary study still has mainly three limitations to be addressed in the future, as follows. (1) Experimental conditions. In this study. to evaluate the performance of a VLC-based image transmission system more precisely, we firstly adopted standard target images for the experiments. These images are quite simple and they can assist us to analyze more effectively and precisely for the test with different receiving areas, test with different angles, and spectrum analysis. However, more results should be provided to further validate the system by using complex environment images with different details and features as cracking, leakage, deformation of underground facilities. (2) Environmental conditions. In this study, we use a dim indoor environment in lab to simulate an underground channel since this study focuses on a preliminary evaluation of the VLC-based image transmission system. However, the actual underground channel is complex, several factors such as effect of background light, light reflection, temperature, dust, fog, and humidity, which can cause light scattering and diffusion in the channel and directly affect the quality of VLC. Therefore, we will test in such environment and consider how to deal with these problems in the future. (3) Deeper analysis. In terms of the low cost and large receiving area, SPs are more suitable for image detection in actual underground environments, but their response time is slow, which will lead to signal distortion in image transmission. Therefore, the measurement of the bandwidth and sensitivity of solar panels are quite important, and this will be addressed in the future.
5. Conclusion and future works
In this study, we have investigated the image signal receiving performance of photodiode (PD) and solar panel (SP) detector in underground facilities visible light communication (VLC) system. These two detectors were placed at a distance of 1.5 m from the optical transmitter for receiving angle and area tests. Through the above tests, some basic conclusions could be obtained. As the receiving angle increased, the image detected and restored by the PD or SP detector would gradually deteriorate in terms of structure, contrast, signal-to-noise ratio, and brightness. Compared with a single detector, double SPs and PDs detectors could increase the brightness and contrast of the image. Finally, in a supplementary experiment, the influence of different optical components on VLC-based image transmission has also been studied. The conclusions could be obtained that the enhanced red optical component could increase the low-frequency signal of the image, the intensity of image become higher while the outlines and edges of the image would be more blurred. Besides, the blue optical component would decrease the low-frequency part of the image. The study on the influence of different optical components on image transmission quality would provide some advantages to improve communication quality in the future.
As the future work, some questions existed should be studied. Firstly, since the scattering and diffusion of white light would cause the attenuation of the optical signal, and therefore, the transmission distance of the image signal was still limited. The long-distance image transmission technology is meaningful in practical applications, and this technology needs to be studied in the future. Secondly, to fully study the impact of different optical elements on VLC, the optical intensity distribution of the LED at the transmitter side and the LED with white, red, green, and blue light at the receiver side should be fully measured. Thirdly, to be applied to actual underground facilities, the standard images we used would not be enough. We will perform the test with different images with different number of details or number of features. This is a very meaningful step to validate and modify our system more quantitatively. Finally, when we use solar panels with much higher sensitivity for image signal detection, several experiments on the effect of distance on transmission performance should be quantitatively carried out.
Acknowledgments
This research was supported in part by Tokyo Gas Co., Ltd. and in part by the Research Institute for Science and Engineering, Waseda University.
Disclosures
The authors declare that there are no conflicts of interest related to this article.
References
1. S. Kim, C. H. Kim, Y. Bae, H. Na, and S. lung, “NDT inspection mobile robot with spiral driven mechanism in pipes,” in Int. Symp. Robotics (ISR, 2013), pp. 24–26.
2. S. Kim, C. H. Kim, Y. Bae, and S. lung, “Development of spiral driven type mobile robot for NDT inspection in small pipes of thermal power plants,” in Int. Conf. Control, Autom. & Syst. (CCAS, 2013), pp. 924–928.
3. T. Nagashima, Y. Tanaka, and S. Ishihara, “Measurement of wireless communication characteristics in sewer pipes for sewer inspection systems using multiple wireless sensor,” in Int. Symp. Pers., Ind. & Mob. Rad. Commu. (PIMRC, 2015), pp. 2055–2060.
4. G. Harutyunyan, K. Blau, M. Hein, and G. Fuch, “Wireless signal transmission through non-metallic pipes with transmission gain,” in Conf. Euro. Micro. (EMC, 2003), pp. 743–746.
5. R. R. Sharma, D. Raunak, and A. Sanganal, “Li-Fi technology transmission of data through light,” Int. J. Adv. Res. Com. Sci. Mana. Stu. 5, 150–154 (2014).
6. A. Sarka, S. Agarwal, and A. Nath, “Li-Fi technology: data transmission through visible light,” Int. J. Adv. Res. Com. Sci. Mana. Stu. 3, 1–12 (2015).
7. R. Karthika and S. Balakrishnan, “Wireless communication using Li-Fi technology,” SSRG-IJECE 2(3), 7–14 (2015). [CrossRef]
8. F. Ebrahimi, Z. Ghassemlooy, and S. Olyaee, “Investigation of a hybrid OFDM-PWM/PPM visible light communications system,” Opt. Commun. 429, 65–71 (2018). [CrossRef]
9. Y. C. Chi, D. H. Hsieh, C. T. Tsai, H. Y. Chen, H. C. Kuo, and G. R. Lin, “450-nm GaN laser diode enables high-speed visible light communication with 9-Gbps QAM-OFDM,” Opt. Express 23(10), 13051–13059 (2015). [CrossRef]
10. W. F. Lin, C. W. Chow, and C. H. Yeh, “Using specific and adaptive arrangement of grid type pilot in channel estimation for white-light LED-based OFDM visible light communication system,” Opt. Commun. 338, 7–10 (2015). [CrossRef]
11. C. W. Chow, C. H. Yeh, Y. F. Liu, P. Y. Huang, and Y. Liu, “Adaptive scheme for maintaining the performance of the in-home white-LED visible light wireless communications using OFDM,” Opt. Commun. 292, 49–52 (2013). [CrossRef]
12. H. M. Oubei, J. R. Duran, B. Janjua, H. Y. Wang, C. T. Tsai, Y. C. Chi, T. K. Ng, H. C. Kuo, J. H. He, M. S. Alouini, G. R. Lin, and S. O. Boon, “4.8 Gbit/s 16-QAM-OFDM transmission based on compact 450-nm laser for underwater wireless optical communication,” Opt. Express 23(18), 23302–23309 (2015). [CrossRef]
13. C. H. Chang, C. Y. Li, H. H. Lu, C. Y. Lin, J. H. Chen, Z. W. Wan, and C. J. Cheng, “A 100-Gb/s multiple-input multiple-output visible laser light communication system,” J. Lightwave Technol. 32(24), 4723–4729 (2014). [CrossRef]
14. M. Kong, B. Sun, R. Sarwar, J. Shen, Y. Chen, F. Qu, J. Han, J. Chen, H. Qin, and J. Xu, “Underwater wireless optical communication using a lens-free solar panel receiver,” Opt. Commun. 426, 94–98 (2018). [CrossRef]
15. W. H. Shin, S. H. Yang, D. H. Kwon, and S. K. Han, “Self-reverse-biased solar panel optical receiver for simultaneous visible light communication and energy harvesting,” Opt. Express 24(22), A1300–1305 (2016). [CrossRef]
16. Z. Wang, D. Tsonev, S. Videv, and H. Haas, “Towards self-powered solar panel receiver for optical wireless communication,” in Proc. Int. Conf. Opt. Net. & Sys. (IEEE, 2014), pp. 3348–3353.
17. Z. Wang, D. Tsonev, S. Videv, and H. Haas, “On the design of a solar-panel receiver for optical wireless communications with simultaneous energy harvesting,” IEEE J. Select. Areas Commun. 33(8), 1612–1623 (2015). [CrossRef]
18. S. Das, E. Poves, J. Fakidis, A. Sparks, S. Videv, and H. Haas, “Towards energy neutral wireless communications: photovoltaic cells to connect remote areas,” Energies 12(19), 3772 (2016). [CrossRef]
19. W. Zhao, M. Kamezaki, K. Yamaguchi, M. Konno, A. Onuki, and S. Sugano, “A preliminary experimental analysis of in-pipe image transmission based on visible light relay communication,” Sensors 19(21), 4760 (2019). [CrossRef]
20. B. Malik and X. Zhang, “Solar panel receiver system implementation for visible light communication,” in Int. Conf. Elec. Circu. Syst. (IEEE, 2016), pp. 502–503.
21. S. Zhang, D. Tsonev, S. Videv, S. Ghosh, G. A. Turnbull, I. D. W. Samuel, and H. Haas, “Organic solar cells as high-speed data detectors for visible light communication,” Optica 2(7), 607–610 (2015). [CrossRef]
22. H.-Y. Wang, J.-T. Wu, C.-W. Chow, Y. Liu, C.-H. Yeh, X.-L. Liao, K.-H. Lin, W.-L. Wu, and Y.-Y. Chen, “Using pre-distorted PAM-4 signal and parallel resistance circuit to enhance the passive solar cell based visible light communication,” Opt. Commun. 407, 245–249 (2018). [CrossRef]
23. J. T. Wu, C. W. Chow, Y. Liu, C. W. Hsu, and C. H. Yeh, “Performance enhancement technique of visible light communications using passive photovoltaic cell,” Opt. Commun. 392, 119–122 (2017). [CrossRef]
24. C. Carrascal, I. Demirkol, and J. Paradells, “On-demand sensor node wake-up using solar panels and visible light communication,” Sensors 16(3), 418–419 (2016). [CrossRef]
25. C. Carrascal, I. Demirkol, and J. Paradells, “A novel wake-up communication system using solar panel and visible light communication,” in Global Commu. Conf. (IEEE, 2014), pp. 461–467.
26. R. Sarwar, B. Sun, M. Kong, T. Ali, C. Yu, B. Cong, and J. Xu, “Visible light communication using a solar-panel receiver,” in Int. Conf. Opt. Commu. & Net. (ICOCN, 2017), pp. 1–3.
27. C. W. Hsu, J. T. Wu, H. Y. Wang, C. W. Chow, C. H. Lee, M. T. Chu, and C. H. Yeh, “Visible light positioning and lighting based on identity positioning and RF carrier allocation technique using a solar cell receiver,” IEEE Photonics J. 8(4), 1–7 (2016). [CrossRef]
28. F. Eren, S. Peeri, M.-Y. Thein, Y. Rzhanov, B. Celikkol, and M. R. Swift, “Position, orientation and velocity detection of unmanned underwater vehicles (UUVs) using an optical detector array,” Sensors 17(8), 1741 (2017). [CrossRef]
29. A. Zisserman, 2D Fourier transforms and applications, http://www.robots.ox.ac.uk/∼az/lectures/ia/lect2.pdf, 2014.
30. R. Wang, Two-Dimensional Fourier Transform, http://fourier.eng.hmc.edu/e101/lectures/Image_Processing/node6.html, 2007.
31. W. Zhao, M. Kamezaki, K. Yoshida, M. Konno, A. Onuki, and S. Sugano, “An automatic tracked robot chain system for gas pipeline inspection and maintenance based on wireless relay communication,” in Int. Conf. Intelligent Robots & Sys. (IEEE/RSJ, 2018), pp. 3978–3983.
32. W. Zhao, M. Kamezaki, K. Yoshida, K. Yamaguchi, M. Konno, A. Onuki, and S. Sugano, “A coordinated wheeled gas pipeline robot chain system based on visible light relay communication and illuminance assessment,” Sensors 19(24), C1 (2019). [CrossRef]
