Borax Detection System in Foodstuffs

Article Study Case BORAX DETECTION SYSTEM IN FOODSTUFFS USING HUSKYLENS CAMERA COLOR RECOGNITION TECHNOLOGY Abstract To prevent borax-contaminated food from being consumed by the public and detect this contamination quickly and effectively, this study proposes the design of an Internet of Things (IoT)-based borax detection tool using the color recognition technology of the HuskyLens Camera.The tool is designed to detect the presence of borax in food with high accuracy in a short period of time.The HuskyLens Camera is used as a borax-sensitive sensor, and the data obtained from this sensor is transmitted through the IoT network to the MIT App Inventor platform. Optimized detection methods and data processing algorithms are used to accurately interpret the sensor results, providing authorities with immediate information on the safety status of the tested food. Experimental test results show that the proposed detector has good performance in detecting borax in various types of food, thus providing an efficient and reliable solution to the problem of food contamination. With the integration of IoT technology, the detector is able to provide real-time information to the authorities, potentially raising awareness about overall food safety. Therefore, the design of this IoT-based borax detector has great potential to address food contamination issues and make a significant contribution to safeguarding public health. Keywords: borax detection, foodstuffs, huskylens, recognition technology INTRODUCTION Exposure to borax in food is a serious food safety issue in Indonesia. Borax is often used as a food additive to preserve and thicken foods such as meatballs, rice cake, crackers, noodles, tofu and children's snacks. The use of borax in food, despite its benefits in processing, can cause acute and chronic poisoning that can potentially endanger consumer health. Several studies have been conducted to identify the presence of borax in street food in various cities in Indonesia, such as Kendari City and Banjarbaru. The results showed that a number of samples of school children's snacks and traditional foods contained borax, suggesting a risk of borax exposure to consumers. In some cases in the field, when collecting samples, officers found food samples that were positive for borax. However, when retested in the laboratory, the food was found to be negative for borax exposure. This is because field officers still use pH paper as a borax test AMELIA ROSE SMITH 1 Amelia Rose Smith indicator in foodstuffs. Some foodstuffs contain natural acids or acids that arise during food processing and storage, especially foodstuffs processed using yeast. Errors in the detection of borax levels will certainly be detrimental to traders. In this context, the development of a borax detector for food is very important. One of the proposed methods is using the color recognition detection method on the HuskyLens Camera. This research aims to develop a borax detection tool that can help identify the presence of borax in food quickly and accurately, so as to improve food safety and protect consumers from exposure to potentially harmful borax. This research presents an innovative solution in the form of a borax detection tool that uses the Color Recognition detection method on the HuskyLens Camera. This camera technology has the ability to recognize colors with a high level of accuracy and can operate in real-time. The color-based detection method was chosen due to the special visual characteristics of borax that can be clearly identified through specific color differences. By applying this technology, it is expected that the detection tool can provide an efficient, fast, and accurate solution in identifying the presence of borax in various types of food ingredients. The implementation of this technology is expected to increase the level of food safety and support efforts to prevent adverse health effects caused by the consumption of products contaminated with borax. Thus, this research is expected to contribute to the development of borax detection technology in foodstuffs, as well as increase understanding of the risk of borax exposure in food in Indonesia. In addition, this research is also expected to provide useful information for related parties in food safety supervision and consumer protection efforts. AMELIA ROSE SMITH 2 Article Study Case METHODS The method used in this research is research and development. Broadly speaking, this research consists of 2 main stages, namely research and development. The research stage includes a literature study related to the research to be carried out. In this research, the literature study includes the limit of borax content in food ingredients, reagents that can be used to detect borax exposure in food ingredients, studying previous research that has been carried out and knowing the right detection method to determine the value of borax exposure in food ingredients. Based on each data and information obtained in the literature study process, a set of theories and possibilities were generated to be tested and applied to this research. The second stage in this research is development. This stage includes planning and managing research so that the results of the research can be proven. Development in this research consists of 3 stages, system planning, system assembly and system testing. Broadly speaking, system planning and assembly consists of 3 parts, namely software, hardware and mechanics. Software planning consists of making system diagrams and making flowcharts to ensure that the system can be built properly. The following is the system diagram and flowchart applied to this system. Firgure 1. Block diagram of system design The system design block diagram in Figure 1 illustrates the system structure consisting of several components with different functions. The main input is provided by the HuskyLens AMELIA ROSE SMITH 3 Amelia Rose Smith Camera, which acts as a sensor to detect color changes in the sample. The detection results from the sensor will be sent to the microcontroller as a process component, then forwarded to the output component. The information received by this camera becomes the basis for decision-making by the system. The Arduino Mega 2560 microcontroller functions as the main process component in the system. It receives detection results from the HuskyLens Camera and performs data processing. This microcontroller becomes the brain of the system that organizes the workflow and produces output according to the information received. The 5V 3A adapter functions as a component that converts AC current into DC and supplies electricity with a voltage of 5 volts and a current of 3 amperes to the microcontroller. Ensuring that the microcontroller has a stable power supply to perform its functions. NodeMCU ESP8266 acts as a connection mediator between the system and the internet. Enables the system to transfer the sensor measurement data to the online database management platform, MIT App Inventor. NodeMCU ESP8266 also plays a role in displaying measurement data in applications developed through the MIT App Inventor platform. The LCD and I2C function as outputs to display the sample readings received from the microcontroller. The LCD provides a direct visual display, while the I2C is used as a communication interface to facilitate data transfer between the microcontroller and LCD. MIT App Inventor is a platform used to create applications that display detection results. This application plays a role in recapitulating and displaying measurement data with a more user-friendly interface on the connected device. The recapitulated measurement data will be collected in a spreadsheet. The system as a whole has an integrated workflow, starting from detection by the camera, data processing by the microcontroller, connection to the internet via NodeMCU ESP8266, to the display of detection results on the LCD and MIT App Inventor application. All components work together to achieve the system's goal of recording, processing, and displaying measurement data effectively. AMELIA ROSE SMITH 4 Article Study Case Start System Initialization Sample Detected HuskyLens Detects changes in sample color Arduino receives the reading data Data is displayed on the LCD Data sent ESP The application displays reading data The reading data is recapitulated in a spreadsheet as a database Figure 2: System flowchart Figure 2 provides a detailed explanation of the designed system workflow, using a flowchart as a visual representation. System initiation is the first step, where all components are activated to prepare for operation. Color detection by the HuskyLens Camera is the next step, where the camera serves as the main sensor to identify color changes in the sample. After color detection is performed, the resultant data is sent to the Arduino Mega 2560 for processing. This microcontroller plays a role in interpreting information from the camera, making decisions based on the data received, and controlling the workflow of the entire system. The NodeMCU ESP8266, as an internet intermediary, is then used to transfer the measurement data to the MIT App Inventor platform which is recapitulated in spreadsheet form, allowing for centralized data storage and wider accessibility. In addition, the detection results are also displayed on an LCD to provide immediate visual feedback to the user. AMELIA ROSE SMITH 5 Amelia Rose Smith The next step includes a more interactive presentation of the detection results through an app developed using MIT App Inventor. The app makes it easier for users to understand and recapitulate the measurement data with a user-friendly interface. Finally, the system completes its process by recapitulating the measurement data in the form of a spreadsheet or online database. The entire workflow reflects the system's well-structured and integrated working principles, ensuring that each component contributes effectively in achieving the system's goals according to the pre-designed scheme. Figure 3: System electrical design Figure 3 displays the electrical schematic of the system in detail in the form of configurations and relationships between components in the system. This schematic has been designed in accordance with the previously compiled block diagram, reflecting the interrelationships between components. This electrical schematic includes all the elements involved in the system operation, starting from the electrical power source with 5V 3A Adapter, HuskyLens Camera, NodeMCU ESP8266, and other components. Each path in the electrical schematic describes in detail how electric current flows between components. The NodeMCU ESP8266 is described as the device responsible for connecting the system to the internet, while the Arduino Mega 2560 is connected to the HuskyLens Camera to receive input from the sensors. The scheme also AMELIA ROSE SMITH 6 Article Study Case includes the use of LCD and I2C as display interfaces to display the detection results and other visual information from the microcontroller. In addition, the connection between the microcontroller and MIT App Inventor, which creates the user application, is described in the schematic to show how the measurement data can be accessed by the user. RESULTS AND DISCUSSION The tests in this study were conducted on 13 different types of samples with different borax levels in each sample. The test results conducted with the tool will be compared with the test results in the laboratory in order to measure the precision level of the tool. The following is a table of the results of the test with the tool and the results of the sample testing with the West Sumatra Food Service laboratory. Table 1. Results of sample testing in the laboratory Code Sample Boraks mg/L 573 Dried Bananas (N) Negative 574 Dried Bananas (F1) Negative 575 Dried Bananas (F2) Negative 576 Maco Dry (F3) Negative 577 Maco dry (F4) Negative 578 Dried Bananas (1B) Positive (50) 579 Dried Bananas (2B) Positive (150) 580 Dried Bananas (3B) Positive (100) 581 Dried Bananas (4B) Positive (50) 582 Maco dry (B1) Positive (100) 583 Maco Dry (B2) Positive (50) 584 Dried Bananas (B3) Positive (100) 585 Dried Bananas (B4) Positive (100) AMELIA ROSE SMITH 7 Amelia Rose Smith Table 2. Sample test results with detection tools Code Sample Boraks mg/L 573 Dried Bananas (N) Negative 574 Dried Bananas (F1) Negative 575 Dried Bananas (F2) Negative 576 Maco Dry (F3) Negative 577 Maco dry (F4) Negative 578 Dried Bananas (1B) Positive (50) 579 Dried Bananas (2B) Positive (100) 580 Dried Bananas (3B) Positive (100) 581 Dried Bananas (4B) Positive (50) 582 Maco dry (B1) Positive (100) 583 Maco Dry (B2) Positive (50) 584 Dried Bananas (B3) Positive (100) 585 Dried Bananas (B4) Positive (100) The test results recorded in Tables 1 and 2 provide a detailed overview of the performance of the detector on various samples. After testing 13 different types of samples, it can be seen that the detector is able to provide highly accurate results, reaching 100% accuracy in 12 test samples and one sample showing an accuracy level of 67%. Overall, it can be concluded that the average or median accuracy of the detection tool reached 97.46%. This figure illustrates the tool's ability to provide consistent and accurate results throughout one test period. It should be noted that the environmental conditions, sample treatment, and reagent concentration levels remained constant throughout the test, so the results can be considered a valid representation of the detector's performance. The test results show a high level of accuracy. However, testing under varying environmental conditions is recommended to test the robustness of the tool. Testing under different conditions can provide further insight into the extent to which the tool can maintain a high level of accuracy in various situations. Thus, the test results provide a solid basis for understanding the performance of the detector in more realistic and diverse scenarios. AMELIA ROSE SMITH 8 Article Study Case Figure 4: Trial sample solution Figure 4 shows the trial sample solution, which is a mixture of reagent and sample solution. It can be seen that the sample solution has a smaller concentration than the reagent solution but has a darker color. This color change is the reference for the HuskyLens Camera to measure the concentration level of borax exposure in the sample solution. Figure 5. (a) Final design of borax detection system in open and (b) closed state. Figure 5 shows the final design of the system that has been designed. Figure (a) shows the device in the open state. There are two parts separated by a partition in the center of the device. The right side of the device is where the electrical components are placed so that they are not damaged during the test and sample detection process. The left side of the tool is a test area that is used as a place for heating the sample solution, the area for detecting the color change of the sample by the sensor and placing the test container. Figure (b) shows the device in closed condition. The lid of the device has a HuskyLens Camera sensor mounted facing the inside of the device so that the sample detection process is carried out when the device is closed. AMELIA ROSE SMITH 9 Amelia Rose Smith CONCLUSION The test results show that from 13 different types of samples, the detection tool can get accurate results with an accuracy rate of 100% for 12 test samples and 1 sample with a test result accuracy rate of 67%. So it can be concluded that the median of the detection accuracy rate of the tool reaches 97.46%. These results were obtained with 1 trial period with the same environmental conditions, sample treatment and reagent concentration levels. To test the durability of the tool, it is necessary to carry out tests with different environmental conditions. AMELIA ROSE SMITH 10 Article Study Case REFERENCES AMELIA ROSE SMITH 11 Amelia Rose Smith AMELIA ROSE SMITH 12 Article Study Case AMELIA ROSE SMITH 13 Amelia Rose Smith AMELIA ROSE SMITH 14