https://indify.co/widgets/live/quotes/hwP6ajJchIIalfbKo0LZ
<aside> <img src="/icons/chat-user_gray.svg" alt="/icons/chat-user_gray.svg" width="40px" />
</aside>
According to a report by the World Health Organization (WHO) in 2016, an estimated 320,000 people died from drowning, making it the third leading cause of unintentional injuries worldwide (Drowning, n.d.). The report also revealed that the majority of victims are children aged 1-4 years old. Current technology employs various approaches to reduce these incidents. Some notable researchers have developed a camera and wearable device that can be attached to the user's body to promptly detect abnormal body functions during a drowning event.
However, using cameras to address this issue has its challenges. Standard cameras can have blind spots when capturing objects in a swimming pool. This is due to obstructions caused by other visitors or the limited range of the camera, which may not cover the entire pool area.
Proposed Idea
Over the past few years, there have been significant developments in camera technology aimed at addressing various limitations in current products. One of the innovations is the 360-degree camera, designed to minimize blind spots and provide a broader perspective. Building on this technology, a new research initiative aims to develop a solution for detecting drowning incidents in swimming pools. The concept involves studying human behavior during moments of panic in the water and using this information to create a safety detection system. With a focus on visitor safety, convenience, and privacy, a 360-degree camera will be installed at the top of the pool to capture all angles. The goal is to swiftly detect any abnormal human motion or activity in the pool.
Testable Questions
| Dependent Variable - Effect: Different ages, Visitors | Independent Variable - Cause: Safety detection, immovable object, random motion, wavelength. | | --- | --- | | Variable Control: 360-degree camera, Swimming pool | Object: Pool's visitors. **** | | General Questions: What indicator mostly happened in the drowning incident in the swimming pool?** | Special Effect:
Main Hypothesis
Experiment Data / Procedure
Fig.01, system's workflow
In support of the smart city concept by integrating every aspect of the city into the innovative system, this research focuses on the safety scenario at the swimming pool. The research data will involve combining a 360-degree camera with a safety detector for drowning cases. The aim is to analyze the abnormal movement and motion of swimmers in water. According to Roy & Srinivasan (2018), individuals in panic situations while drowning can only stay afloat for 1-2 minutes. Therefore, it is important to have faster detection to address this issue. The proposed system will involve three steps: Recording, Post-Editing, and Technical Analysis.
During the recording step, the 360-degree camera will be placed at a top-view angle to provide a more comprehensive visual of the swimming pool while minimizing blind spots caused by a high number of people. The resulting video format will be different due to the merging of the two camera feeds into one view.
The post-editing step involves converting the curved video format from the 360-degree camera into a 2D flat video for analysis. This process aims to address accuracy issues that may be caused by different lighting conditions. Color grading treatment is included in this step to ensure the video is optimized for technical analysis.
The drowning detection system will focus on the technical analysis step. A drowning simulation will be conducted using three different groups - children, adults, and the elderly. By analyzing the data, the research aims to identify patterns of drowning through motion, movement, and water waves. The system will then use mathematical formulas to interpret the data and establish a suitable pattern for adoption. The expected result is a drowning detection system that effectively utilizes the 360-degree camera to analyze data based on the hypothesis that immovable objects and random motion create water wave patterns indicative of drowning.
Assumed Data and Data Analysis
(How to get the data: 1. Simulation Test, 2. Video review of drowning case.)
TQ-1. Assuming the time of people drowning in the swimming pool, calculating starts from the object not moving and makes a panic motion on the water to head entirely underwater.
H0: No correlation of the different levels of ages to the time of camera detecting drowning scenario.
H1: There is a correlation between the different levels of age to the time of the camera detecting the drowning scenario.
| ** | Elderly/s | Child/s | Adult/s |
|---|---|---|---|
| ** | 0.5 | 0.5 | 1 |
| ** | 0.3 | 0.5 | 1 |
| ** | 0.001 | 3 | 2 |
| ** | 0.09 | 4 | 1.5 |
| ** | 0.9 | 0.5 | 1.1 |
| ** | 0.2 | 0.4 | 1.2 |
| ** | 0.5 | 0.1 | 1.3 |
| ** | 0.6 | 0.2 | 0.01 |
| ** | 0.007 | 0.5 | 0.5 |
| ** | 0.01 | 0.2 | 0.2 |
| Mean | 0.31 | 0.99 | 0.98 |
| SD | 0.31 | 1.35 | 0.60 |
Chi-square
The chi-square analysis reveals that the p-value of the table is 0.9795, which is based on the standard p-value of 0.05, and the chi-square value is 7.944. With a significance value of 0. 9795 being greater than 0.05, the Null hypothesis is accepted, indicating no correlation between the immovable object and random motion, as represented by elderly, children, and adults, in the context of the camera's speed in detecting a drowning scenario. However, the analysis also shows no significant relationship between the independent and dependent variables.
Furthermore, the "Pairwise Comparison Analysis" was conducted to scrutinize the three different age groups (immovable objects with random motion) to discern which object affects the camera's speed in detecting drowning. The comparison indicates that none of the age groups are significant, as the p-values for each category are all 1.0000. It can be concluded that the varying age groups of drowning objects (people) do not significantly affect how fast the camera detects the drowning object. Thus, it can be inferred that humans exhibit similar levels of spontaneous panic sensitivity across all age groups.
ANOVA
To obtain the most accurate measurement of the table data, this study also explores the use of ANOVA. The experiment focused on a single object group, specifically the 360-degree camera detection as the dependent variable. The average values in the table for each category are 0.31, 0.99, and 0.98 for the elderly, children, and adults, respectively. The P-value reflects the same result as the chi-square analysis, indicating an insignificant value with P=0.1266, which is less than P=0.05 (F2.18=2.323, P<0.05). This is related to the null hypothesis, suggesting that the stationary object with different age groups does not significantly affect the speed of object detection by the camera.
In conclusion, after conducting two analyses, it was found that there is no significant evidence to reject the null hypothesis, indicating no correlation between the independent and dependent variables.
TQ-2. Assuming data of water waves made in the panic situation, measuring the average data of the wavelength started from the first panic motion to the head completely underwater.
H0: No correlation of different ages with the wavelength they produced.
H1: There is a correlation between different ages and the wavelength they produced.
| Elderly/cm | Child/cm | Adult/cm |
|---|---|---|
| 5 | 2.5 | 7.8 |
| 6.1 | 3.3 | 8.9 |
| 6 | 5.1 | 5.4 |
| 8.1 | 3 | 7 |
| 5 | 5 | 6 |
| 5.5 | 2 | 8 |
| 4.1 | 1.2 | 5.6 |
| 6 | 5 | 8 |
| 5 | 1 | 7.6 |
| 5.1 | 2 | 8.5 |
Upon analyzing the Chi-square table, it was found that the p-value is 0.9812 at a significance level of 0.05, and the chi-square value is 7.818. Since the p-value of 0.9812 is greater than 0.05, the Null hypothesis is accepted, suggesting no correlation between the wavelength and the different age levels of users. It is also noted that the result does not show a significant relationship between the independent and dependent variables.
Furthermore, the "Pairwise Comparison Analysis" was utilized to examine three different age levels and identify any divergences. The analysis revealed that none of the age categories are significant, as each one has a p-value of 1.0000. Consequently, it can be inferred that varying age levels do not yield significant differences in wavelengths, indicating that humans exhibit similar responses in panic conditions.
TQ-3. Assuming the number of people coming based on the visitor data of the swimming pool with different kinds of safety tools.
H0: No correlation of safety tools with the number of visitors.
H1: There is a correlation between safety tools with the number of visitors.
| Day | Guard | Live-camera | Both |
|---|---|---|---|
| 1 | 300 | 200 | 500 |
| 2 | 250 | 250 | 350 |
| 3 | 450 | 150 | 600 |
| 4 | 200 | 100 | 700 |
| 5 | 300 | 100 | 800 |
The data analysis reveals a p-value of 0.0000 at a significance level of 0.05, along with a chi-square value of 360.355. The significance analysis indicates that 0.0000 < 0.05, leading to the acceptance of hypothesis one. This suggests a correlation between safety tools and the number of visitors. Furthermore, there is a significant relationship between the independent variable and the dependent variable.
Given that the independent variable comprises three tool categories, a pairwise comparison is necessary to identify the differing tools. This comparison shows a statically significant result for the 2:3 pair (p = 0.0034, p<0.05), despite the higher chi-square result. Consequently, this suggests a difference in the number of visitors between the live-camera tool and both guard and camera tools. However, the 1:2 (p = 0.5014) and 1:3 (p = 0.6440) pairs do not show significance based on the (p<0.05) threshold.
Conclusions and Statements
The research findings indicate that age does not significantly impact how the camera detects scenarios involving stationary objects with motion or the wavelength produced by moving objects representing people. The analysis suggests that the presence of safety tools, such as guards, live cameras, or both, does affect the number of visitors, with the most significant impact observed when both live cameras and guards are present at the swimming pool.
In relation to the main hypotheses MH1-2 and MH3-4, the results show that body motion and movement can serve as indicators of drowning but do not significantly affect how the camera detects scenarios. Additionally, it is evident that people prefer the presence of both live cameras and guards at the swimming pool rather than just one of these safety measures. Privacy does not appear to be a primary concern for visitors as long as it remains within the boundaries of safety.
Reference
Drowning. (n.d.). Retrieved June 26, 2020, from https://www.who.int/news-room/fact-sheets/detail/drowning