Trang 1 MINISTRY OF EDUCATION AND TRAINING HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY AND EDUCATION FACULTY FOR HIGH QUALITY TRAINING DESIGN OF AN AUTOMATED BILLING SYSTEM ADVISOR : TRU
TOPIC OVERVIEW
Introduction
The fourth industrial revolution, which emerged in the latter half of the 20th century, has significantly advanced scientific and technological progress, reshaping global economies, including Vietnam's This revolution has enhanced production capabilities, increased labor productivity, and transformed economic structures worldwide Its practices aim to meet humanity's expanding material and spiritual needs, profoundly influencing all aspects of society A thriving economy relies on diverse industries and service sectors essential for goods production and service delivery Vietnam's economy is experiencing rapid growth, with improvements in the supply chain and a rise in automation systems, which are crucial for business development Additionally, shifts in consumer preferences and purchasing behaviors are reshaping the Vietnamese market, encouraging public contributions to national growth.
2 recent few years People may now purchase food, beverages, and other necessities from billing machines located all across the country
A billing machine is a device that dispenses goods such as food, drinks, and lottery tickets in exchange for cash or cards The first modern coin-operated billing machines were introduced in London in the 1880s, with Percival Everitt creating a popular model in 1883 for dispensing postcards and envelopes at train stations In 1887, the Sweetmeat Automated Delivery Company became the first business dedicated to the installation and maintenance of these machines By 1893, German chocolatier Stollwerck had integrated its products into 15,000 billing machines, expanding to offer a variety of items including chocolate, cigarettes, and chewing gum.
Despite the widespread use of billing machines globally, significant limitations remain, as highlighted in Nguyen Ngoc Phuong's article in the Journal of Science and Technology Education These machines often cannot accept credit card payments, have a limited product range, are vulnerable to theft, and are restricted to public locations To address these challenges, our team believes it is essential to develop a new billing machine from scratch, rather than modifying existing models A notable example of innovation in this field is Amazon Go, which showcases a modern approach to vending technology.
In 2016, Seattle witnessed the introduction of Alibaba Group's innovative "Taocafe" in Hangzhou in July 2017 This intelligent store leverages advanced technologies, including embedded systems, the Internet of Things, and artificial intelligence It is suggested that a sales system be created that mimics a traditional store experience while eliminating the need for in-person staff.
To enhance productivity and minimize time and financial losses for workers, it is essential to implement a streamlined payment system This approach aims to simplify product management while achieving significant efficiency gains Furthermore, developing a sales system akin to a retail model, which operates independently of a payment crew, is highly recommended.
Aim and scope
- Design of an automated billing system
- Identify products 2 times: first time is barcode and second time is AI.
Methodology
- Research and study the billing system
- System test, integrated circuit and evaluation
Outline of report
The main content have 5 chapters:
This article begins with Chapter 1: INTRODUCTION, which outlines the research objectives, methods, and limitations Chapter 2: THEORETICAL BASIS provides an overview of the current research landscape, including relevant services In Chapter 3: SYSTEM IMPLEMENTATION, we present a comprehensive model of the system, detailing its components and the design of each block along with the devices utilized Chapter 4: EXPERIMENT RESULTS showcases the outcomes of the system model construction Finally, Chapter 5 focuses on the SELF-ASSESSMENT of the subject implementation.
CONCLUSION: drawing conclusions, strengths and weaknesses and development direction of the model
THEORETICAL BASIC
Billing system overview
Larger populations correlate with faster market growth and increased consumer purchasing power, as they encompass a greater total number of individuals In Vietnam, there is a promising opportunity for revenue generation through the sale of automated equipment However, the concept of automated sales is still in its early stages, with current machines representing the most basic form of this system The implementation of this system will likely determine its impact on sales, influencing whether the effect is positive or negative.
-The simplicity is easy to use, preserving products quite well
- Continuous operation of the day to ensure customers can buy in any time frame
-The price and clear information
-Do not take up much space and do not take much labor costs
-Technical errors during operation such as not receiving money, unable to offer the product
2.1.2 Billing system in the world
RFID technology is revolutionizing billing machines by integrating RFID tags on products, allowing an automatic system to transport items to the payment area using an RFID reader Customers can conveniently pay for their purchases with bank cards or RFID membership This innovative system enhances the payment process compared to traditional barcode scanning by eliminating restrictions on product types and accelerating transactions without increasing labor costs.
The IoT billing machine application revolutionizes the shopping experience by allowing customers to scan a QR code with a smartphone before selecting products Utilizing advanced cameras and sensors, the system employs machine learning and facial recognition algorithms to identify purchased items, enabling automatic online payments without the need to wait in line This innovative approach not only enhances convenience for customers but also necessitates a considerable investment in technology.
Although these systems have excellent automatic sales benefits, widespread application in practise would require a lengthy rescue process and significant investment funds.
Process of billing system
To initiate the process, barcodes will be attached to physical items for scanning When a customer purchases an item, it will be marked on the conveyor belt, where a barcode reader will scan it as it passes This scanning occurs with each purchase, allowing the scanner to read the barcode information and transmit it to the system for processing The product will then move through the system while a camera continuously monitors it Once the scanning is complete, the camera will analyze the product's image to identify it If the product's shape matches the product code, the corresponding price information will be displayed on the screen.
Customers will use RFID cards for payment, allowing the barrier to open once the transaction is completed, enabling them to access their purchased products.
There is a camera combined with a barcode scanner because avoiding the following cases:
- Case 1: Customers tear the barcode of low -value products affixed to high -value products wrongly fraud to redeem
-Case 2: Customers tear off the barcode of the product and then pay, resulting in cases where the product does not have a barcode
To effectively address the issue, we develop a strategy that involves identifying products through the simultaneous analysis of camera data and object cost scanning The system restricts barcode usage if either condition is unmet or if there is a mismatch between the scanner and camera information Furthermore, mathematical calculations and the operation of conveyors will be hindered under these circumstances.
Hardware design
The Arduino Nano is a lightweight and compact microcontroller board, weighing just 7 grams and measuring between 1.8 to 4.5 cm Ideal for those who prefer small designs, the Arduino Nano offers the same functionality, power, hardware, and usability as larger Arduino models, making it an excellent choice for various projects.
The Arduino Nano 3.x utilizes the ATmega328 microcontroller, while the Arduino Nano 2.x is based on the ATmega168 Although it operates similarly to the Arduino Duemilanove, the Nano features a different form factor, lacking a DC power jack and instead employing a USB Mini connection for power and programming.
B cable to operate The Arduino Nano is produced and designed all over the world
Figure 2: Arduino Nano Pin Diagram [4]
Table 1: Explanation of Configuration Pin the Arduino Nano Pin Diagram [5]
Pin Number Pin Name Type Description
3, 28 RESET Input Reset (active low)
+5V output (from on-board regulator) or +5V (input from external power supply)
Figure 3: Schematic of Arduino Nano
-Microcontroller: ATmega328P – 8-bit AVR family microcontroller
-Recommended Input Voltage for Vin pin: 7-12V
-Digital I/O Pins: 14 (Out of which 6 provide PWM output)
-DC Current on I/O Pins: 40 mA
-DC Current on 3.3V Pin: 50 mA
-Flash Memory: 32 KB (2 KB is used for Bootloader)
The output driving capacity of the A4988 stepper motor driver is up to 35V and 2A You can control a bipolar stepper motor like this
The stepper motor operates quietly due to controlled output current, eliminating the resonance common in unregulated driver systems Additionally, the driver features built-in translation, simplifying control to just two pins: one for step control and another for direction regulation.
The driver offers five different step resolutions: full-step, half-step, quarter-step, eighth- step, and sixteenth-step as per Table 2
Table 2: Rotation sequence input in stepper motor drive [8]
Microstep Resolution MS1 MS2 MS3
In order to ensure reliable operation, the driver has additional features such as under- voltage, shoot-through, short circuit, overcurrent, and thermal protection
Table 3: Explanation of Configuration Pin in the A4988 Pin Diagram
The Power State pin of the A4988 driver is a low-active input that is typically pulled LOW, keeping the driver enabled To disable the driver, you must pull this pin HIGH, making it ideal for implementing an emergency stop or shutdown mechanism.
Power State is also an active low input Every STEP input is ignored while this pin is pulled LOW By returning the internal translator to a predetermined "home" state, it
12 also resets the driver Home state, which varies depending on microstep resolution, is essentially the starting position from which the motor moves
The Power State pin is an active low input that triggers the driver to enter sleep mode when pulled LOW, significantly reducing power consumption This feature is especially beneficial for conserving energy when the motor is idle.
The motor's microsteps are regulated by input signals, with each HIGH pulse on the designated pin corresponding to the selected number of microsteps As the pulse frequency increases, the motor's speed also rises, enabling faster rotation.
The motor's spinning direction is controlled by input The motor rotates clockwise when it is pulled HIGH, and counterclockwise when it is pulled LOW
9 GND PWR power the internal logic circuitry, which can be powered by voltages of 3 to 5.5 volts
15 GND PWR provide the motor with power, which can be between
The MFRC522 controller from NXP Semiconductors is the foundation of the 13.56MHz RFID module known as the RC522
The module includes an RFID card and key fob, supporting I2C, SPI, and UART interfaces It is commonly used for person and object identification in attendance systems and various other applications.
Figure 8: Module RF RC522 Pin Diagram [8]
Table 4: Explanation of Configuration in the Module RF RC522 Pin Diagram [12]
1 VCC Used to Power the module, typically 3.3V is used
2 RST Reset pin – used to reset or power down the module
3 Ground Connected to Ground of system
4 IRQ Interrupt pin – used to wake up the module when a device comes into range
5 MISO/SCL/Tx MISO pin when used for SPI communication, acts as SCL for I2c and Tx for UART
6 MOSI Master out slave in pin for SPI communication
7 SCK Serial Clock pin – used to provide clock source
8 SS/SDA/Rx Acts as Serial input (SS) for SPI communication, SDA for IIC and Rx during UART
Figure 9: Schematic of Module RF RC522
-Frequency Range: 13.56 MHz ISM Band
-Min Current (Power down): 10àA
Liquid Crystal Display, commonly referred to as LCD, is a type of electronic display module found in various devices such as mobile phones, calculators, computers, and televisions This technology is distinct from OLED (organic light-emitting diode) displays, offering a versatile solution for numerous electronic applications.
Light-emitting diodes (LEDs) predominantly favor this display form, whether featuring seven segments or more Key benefits of this module include its affordability, ease of programming, and capability to showcase endless personalized characters and animations Additionally, it offers limitless potential for displaying custom features and various animated elements.
Table 5: Explanation of Configuration in the LCD 16x2 Pin Diagram [14]
Pin 1 (Ground/Source Pin): The display's GND pin connects to the microcontroller's
GND terminal or the power supply
Pin 2 (VCC/Source Pin): This is the display's voltage supply pin, which is used to attach the power source's supply pin
Pin 3 (V0/VEE/Control Pin): This pin, which is used to connect a movable POT that can supply 0 to 5V, controls the difference of the display
Pin 4 (Register Select/Control Pin): Used to connect a microcontroller unit pin, this pin switches between command and data registers It receives either 0 or 1 (where 0 is data mode and 1 is command mode)
Pin 5 (Read/Write/Control Pin): Connected to a microcontroller unit pin to receive either 0 or 1 (where 0 indicates a write operation and 1 indicates a read operation), this pin toggles the display between read and writes operations Pin 6 (Enable/Control Pin): This pin is connected to the microcontroller unit and is always held high It should be held high to execute the Read/Write process Pin 7-
Data pins are essential for transmitting information to the display, utilizing two-wire configurations like 4-wire and 8-wire modes for connections In 4-wire mode, only four pins (0 to 3) are connected to the microcontroller unit, while in 8-wire mode, eight pins (0 to 7) are utilized.
Pin 15 (+Ve pin of the LED): This pin is connected to +5V
Pin 16 (-Ve pin of the LED): This pin is connected to GND
-The operating voltage of this LCD is 4.7V-5.3V
-It includes two rows where each row can produce 16-characters
-The utilization of current is 1mA with no backlight
-Every character can be built with a 5×8 pixel box
-The alphanumeric LCDs alphabets & numbers
-Is display can work on two modes like 4-bit & 8-bit
-These are obtainable in Blue & Green Backlight
-It displays a few custom generated characters
With 200 steps per revolution, this stepping motor has a 1.8° step angle With 1.7 A at 2.8
The motor can achieve a holding torque of 3.7 kg-cm for each phase, with the black and green wires linked to one coil and the red and blue wires connected to the other coil, both ending in bare leads To effectively control the motor, a pair of suitable H-bridges, one for each coil, is required.
Besides the above hardware devices, there are other devices in the system, such as:
-Barcode scanner YHD 1100: Used to scan product barcodes
- Waterproof, dustproof, up to IP54 industrial level
- Manual trigger, continuous scan, or auto-sense (as options), very comfortable to use
- Fuzzy Information Processing System with Quick Decoding Capability
- Up to 10,000 hours of laser life, with a French push key with a life of 0.33 million presses
- Support multilingual language keyboard RS232, USB, and KB multi-interface support
- Ergonomic designs provide comfortable use and boost productivity
- Compatible software includes Android, Linux, Windows 98, and all other systems, plug and play without a driver
- Camera: Used to identify product images
Figure 14: Camera Full HD 1080p webcam ACOME AWC11
Specifications of the Full HD 1080p webcam ACOME AWC11 - Full HD 1080P 2MP resolution and a 1/2.9 CMOS sensor for sharper and more fluid pictures
Flexible camera rotates 360 degrees horizontally and tilts 15 degrees, noise- cancelling microphone for smooth talks, USB port connection compatible with Windows/Android/iOS/Linux platforms
Useful for online education, livestreaming, video calling, identification, surveillance, and tracking
Three ways to utilize Mount on computer display, stand, or tabletop
Model Yolo
You Only Look Once (Yolo) is one of the fastest object recognition algorithms available today While it may not be the most accurate method, it remains popular for real-world applications where precision is not the highest priority Yolo has undergone three iterations so far, and this article will guide you on how to train Yolo version.
Object detection involves not only classifying items within an image but also identifying their locations Traditional methods prior to YOLOv1 typically utilized a two-step process: first, a sliding window or algorithm would select potential areas that might contain objects, and then these areas would be classified to determine the object's class However, these approaches are labor-intensive and fragmented, complicating performance optimization In contrast, YOLOv1 treats object detection as a regression problem, employing a combination of convolutional, pooling, and fully connected layers to produce outputs directly from the input image This architecture is optimized for GPU processing with a single forward pass, enabling remarkably fast detection speeds.
In late 2016, Joseph Redmon and Ali Farhadi introduced YOLOv2, also known as YOLO9000, which was set to be released at the 2017 CVPR This version significantly improves upon its predecessor by offering enhanced speed and advanced capabilities, effectively addressing the challenges encountered with YOLOv1 and keeping pace with the faster R-CNN technique that utilizes a Region Proposal Network.
Using anchorbox architecture to make predictions
K-mean clustering for anchor selection
YOLO9000 introduces a technique for integrating supplementary datasets with ImageNet to expand class identification The author creates a directed graph known as WordTree to merge labels from the ImageNet dataset, which contains 1,000 classes, with COCO/PASCAL, featuring 100 classes By utilizing WordNet to connect these classes, the author enhances the network's ability to detect significant classes effectively.
YOLOv3 has an architecture identical to YOLOv2 The author has updated YOLOv2 with latest changes to generate YOLOv3 These enhancements consist of:
Logistic regression for confidence score: Using logistic regression, YOLOv3 predicts bounding box confidence (contains objects or not)
YOLOv3 enhances object categorization by replacing the traditional softmax function with discrete logistic classifiers This approach is particularly advantageous for scenarios where labels are not mutually exclusive, allowing for the identification of multiple classes simultaneously For example, in a task that requires detecting both a "dog" and a "Phu Quoc dog," the latter inherently falls under the broader category of "dog."
"dog" class; hence, softmax is inappropriate here
New Backbone – Darknet 53: Redesigned backbone using leftover blocks
Multi-scale prediction: Feature Pyramid Networks (FPN) architecture is used by
YOLOv3 to produce predictions based on different sizes of feature maps This allows YOLOv3 to use feature maps with varying degrees of coarseness and fineness for prediction
YOLOv3 employs skip-layer concatenation, which enhances the model's ability to predict small objects by establishing relationships between prediction layers This process involves upsampling the prediction classes from the later layers and concatenating them with those from earlier layers, resulting in improved accuracy and performance in detecting tiny items.
Conclusion: YOLO's difficulty with real-time object identification has gained widespread notoriety However, since YOLOv3, the original creator of YOLO, Joseph Redmon, has
The development of computer vision technologies, including YOLO, has faced challenges due to concerns over potential misuse, such as military applications and privacy issues Despite these worries, the field continues to evolve, with updates like YOLOv4 and YOLOv5 being released by various contributors Joseph Redmon has also shared insights on this topic through social media, highlighting the ongoing advancements in the technology.
Figure 15:Flowchart of recognition operation using yolo
Connect web camera to receive image data
DESIGN OF AUTOMATED BILLING SYSTEM
Block diagram
The system consists of two main components:
The first layer of the system is a lower-level control layer managed by an Arduino microcontroller This layer, which operates beneath the second layer, is tasked with receiving signals from RFID cards Once the signal is received, it is forwarded to the control unit, which oversees the operation of various system mechanisms, including conveyors, gates, and display panels.
Figure 16: Block diagram of billing system
Each block is tasked with completing its unique assignments, functioning independently within the system Like all systems, it requires a reliable energy source to maintain its operations.
For the system to function optimally, it is crucial to ensure it has a reliable energy source This principle applies universally, including to this particular system, which requires consistent energy input to maintain its performance.
Once the previous step is completed, the power block will initiate the delivery of 12 volts DC to the system's devices.
An automated identification system utilizing radio frequency identification (RFID) technology enables remote data storage through RFID tags, offering significant advantages The RFID block transmits scan data from an RF card to the Arduino at speeds up to 10 Mbps via the SPI interface The Arduino microcontroller processes the incoming information and generates the appropriate output signal for the billing system.
The A4988 control power block is essential for managing the operation of stepper motors within a system, placing the responsibility on the user It can effectively control motors with a maximum operating voltage of 35 volts and a current capacity of 2 amperes, provided the motors are designed appropriately.
The step motor within the executive block activates as soon as data is received from the A4988, driving the conveyor belt to facilitate the smooth and stable transportation of goods This motor is housed in a dedicated compartment within the conveyor system, ensuring efficient operation and enhanced logistics.
The Servo 995 functions as an engine to simulate a door used for picking up goods, successfully mimicking the opening and closing motion essential for this operation The door remains closed until the customer completes the payment for the item, ensuring they cannot claim possession until the transaction is finalized We appreciate your patience as we await the customer's payment before activating this engine.
The user interface aims to simplify the management of login features, consumer cards, and individual items, which include details such as name, item code, and unit price This streamlined approach enhances the overall user experience by making it easier to access and manage essential information.
The procedure allows for the addition of products either manually or automatically, based on customer preference In automatic mode, the scanner matches the scanned code with the data warehouse to determine unit pricing and compute the total cost For manual additions, the scanner performs the same matching process Customers receive a detailed breakdown of the total cost via the LCD display and user interface Upon payment, the displayed amount resets to zero, funds are withdrawn from the customer's card, the previous transaction is canceled, and the system is ready for the next order.
Figure 17: Flowchart of system operation
Connection Diagram
Figure 18: Connection diagram of billing system
The system's connection diagram highlights the role of the Arduino Nano as the command and control node, interfacing with various electrical components As illustrated in Figure 20, the Arduino Nano will connect to the RF RC522 module, with the connection process commencing shortly Additionally, a computer will be linked to the Arduino Nano, but only after it establishes connections with a camera and scanner to gather product data through scanning and recognition This sequence ensures that data is processed before the computer communicates with the Arduino Nano.
31 is hooked up to the personal computer (PC) The Arduino Nano will be connected to the personal computer when this connection has been made (PC)
The Arduino Nano plays a crucial role in managing the system by sending signals to the LCD to display prices and notifying bartenders when a customer has completed their payment, allowing them to open the bar This process is initiated after the customer finishes paying and the LCD is updated accordingly Additionally, the A4988 driver controls a stepper motor, which is connected to the Arduino, enabling seamless communication between the two technologies The stepper motor is then attached to a conveyor belt, facilitating the transportation of items within the system.
Figure 19 : Flowchart of operation on the control circuit
Note: BTC is Button for Conveyor and BTB is Button for Barrier start
3.2.1 Connection diagram between arduino nano and Module RF RC522
Figure 20: Connection diagram between arduino nano and Module RF RC522[20]
Red wire is Vcc pin used to power 3.3V the module,
The black wire serves as the ground pin, connecting to the grounding point and featuring a 9-pin configuration with a reset active low The yellow wire acts as the reset pin for the module and can also power it off Pins 10, 11, 12, and 13 are designated for SDA, MOSI, MISO, and SCK, respectively In SPI communication, these pins function as MISO (Master In Slave Out), MOSI (Master Out Slave In), SCK (Serial Clock), and SDA (Serial Data Line) for IIC communication Additionally, they perform as SCK and MOSI in SPI setups.
Table 6: Connection pins between arduino nano and Module RF RC522
3.2.2 Connection diagram between arduino nano and LCD 16x2
Figure 21: Connection diagram between arduino nano and LCD I2C 16x2[21]
In Figure 17, two blue wires are connected to the Arduino Nano, facilitating the transmission of an analog signal to the LCD These cables are linked to the A4 and A5 pins, which serve as the analog input channels for the device.
In close proximity, we can identify a black wire, which signifies the ground, and a red wire, indicating the power supply.
Table 7: Connection pins between arduino nano and LCD 16x2
Figure 22: Schematic of Arduino nano and LCD
3.2.3 Connection diagram between arduino nano and servo 995
Figure 23: Connection diagram between arduino nano and servo 995 [22]
Connecting the Servo 995 to the Arduino Nano is straightforward, requiring only three wires: yellow for data, black for ground, and red for power The bright yellow wire transmits data to pin 27 of the Arduino Nano, which outputs 5V The black wire connects to the ground, while the red wire supplies power Additionally, the power and ground connections from the breadboard are linked to the corresponding pins on the Arduino Nano, ensuring a simple and efficient setup.
Table 8: Connection pins between arduino nano and servo 995
3.2.4 Connection diagram between arduino nano and A4988 driver
Figure 24: Connection diagram between arduino nano and A4988
The STEP pin on the A4988, connected to pin 5 of the Arduino Nano as a digital output, controls the motor's microsteps Each HIGH pulse sent to this pin drives the motor according to the selected microstep configuration As the motor accelerates, the pulse frequency increases, allowing for smoother and faster operation.
The 2 pin serves the same purpose as the 5 pin by connecting to the DIR pin, which acts as a control input for motor direction When the control input is pulled HIGH, the motor spins counterclockwise, while pulling it LOW causes the motor to rotate clockwise.
Table 9: Connection pins between arduino nano and A4988
Figure 25: Schematic of Arduino nano and A4988
EXPERIMENT RESULTS
Overview of system
The fully automated billing machine, once constructed, will enable users to scan barcodes and arrange items on a moving conveyor belt for seamless purchasing Upon scanning a product's barcode, the sales interface will update accordingly, while a camera system at the end of the conveyor will verify that the item's information is accurate If everything checks out, the customer can swipe their credit card to complete the transaction, granting access to the warehouse for item retrieval In cases where a product is incorrectly detected after scanning, the customer will be prompted to address the issue.
40 swipe their payment card Currently, the staff must manually review the order and total amount in the software.
Operation of system
The program responsible for order processing, system operation, commodity codes, and user management
Upon launching the software for the first time, you will be prompted to log into your account Completing this step is essential for continued access to the program.
Next when we log in, a successful login window will appear on the screen
When accessing the payment interface, users will see a payment window featuring a detailed list of their order, including the order number, item name, product code, and unit price on the left side The right side displays the product image for easy recognition The central focus of this window is the payment section, which clearly indicates the status of the recognized products for a smooth transaction experience.
Figure 29: Interface for processing, payment of orders and application of the system
SQLite data encompasses crucial information such as warehouse details, user accounts, and login credentials It includes client payment card information, login details for company owners, and commodity codes related to the facility's data storage.
The table contains information pertaining to the following several categories: barcodes to identify the product, as well as monetary reimbursement for each individual item
The table provides details about the consumer's payment card, including the total number of available cards and the remaining balance on each card.
The user interface for managing consumer information allows for the registration of a new card through a simple swipe, along with the creation of a username and the option to add funds Users can opt to skip card registration and proceed with their transactions After scanning the cards and entering the desired amount, users can confidently add money, ensuring the correct amount is allocated to each card.
Figure 33: Customer information management interface
A table featuring fields for the collection of information about customers, such as their retail establishment login name and password
The labeling position involves monitoring products on a conveyor belt, where a camera system with image recognition software identifies items as they reach the end Upon halting for inspection, if both the barcode and product image contain errors, the system alerts the seller, enabling them to assess the order for the customer's benefit Conversely, if the information on the screen matches the barcode data, the system will display the barcode correctly.
To ensure the accuracy of the product identification model, both the scan code and customer purchase information were utilized This approach allowed the model to consistently maintain its precision By focusing on these data aspects, we achieved our goal efficiently and effectively The information captured not only reflects the total purchase price but also details the quantity of each item and the overall number of goods acquired.
The total order amount is displayed on the LCD screen, which also features the store's accessible payment interface This transparency ensures that customers can confidently complete their purchases, fully aware of the value of the items they are buying.
4.2.4 The role of the barie
Barie bars are intended to open and close so that clients may get their purchases upon payment After a certain amount of time, the barrier automatically lowers
An open bar operates on a payment system where customers gain access only after completing their purchase Once payment is made, a barrier lifts, allowing entry, and after a set duration, the items will start to accumulate.
This is the current state of the barrier that is being constructed It prevents the product from being stolen without payment, which is its primary purpose
CONCLUSION
Result
The project implementation deepened our group's understanding and provided valuable lessons learned, while also offering insights into the feasibility of applying these findings in various contexts.
- During the research process, we have applied the knowledge we have learned and learned more new knowledge to complete the set requirements
- The product works 90% of the required
- The circuit works stably and accurately when switched on and off many times
The circuit is designed to be simple, compact, and easy to assemble, featuring minimal components that are readily available on the market Its automatic and secure closed payment model ensures a straightforward and efficient payment process, ultimately saving labor and enhancing user experience.
Development
While the system has been effective, the camera's item recognition mechanism is hindered by the computer's inadequate configuration Upgrading to more advanced hardware is essential to eliminate delays during product scanning.
Products in the list are few, try to add more products to make the product system more diverse
Combined with the product self-packing system to produce the product that is packed in the box
Develop more payment possibilities, can pay in cash using artificial intelligence for money recognition
[1] Hui, Mary“Vending machine for homeless just launched in U.K., to debut in U.S cities” Chicago Tribune
[2] Chandler, Alfred (2004) Scale and scope: the dynamics of industrial capitalism Cambridge, Massachusetts: Belknap Press
The article discusses an innovative self-service vending system designed for smart retail, developed by Kun Xia, Hongliang Fan, Jianguang Huang, Hanyu Wang, Junxue Ren, Qin Jian, and Dafang Wei from the Department of Electrical Engineering at the University of Shanghai for Science and Technology This intelligent system aims to enhance the shopping experience by integrating advanced technology to streamline transactions and improve customer convenience in retail environments.
The article discusses the design of fire protection systems for electrical power distribution panels at Lippo Plaza Sidoarjo, utilizing Arduino Nano technology It emphasizes the importance of implementing effective safety measures to prevent fire hazards in electrical installations The study, conducted by Achmad Sholih as part of the Electrical Engineering program at Muhammadiyah University of Sidoarjo, highlights innovative approaches to enhance fire safety in commercial buildings.
[7][8][9][10] Anmol Singh Shekhawat "Design and Control of Two-wheeled Self- BalancingRobot using Arduino"Electrical and Electronics Engineering IET, JK Lakshmipat University,Jaipur, India
[11][12][13] Nanshi Bhuwalka and Rahul P B “Intelligent Traffic Control System” Visvesvaraya Technological UniversityBelgaum, Karnataka
[14][15] Sunkaraneni Sumanth"IoT BASED EVENT MANAGEMENT SYSTEM USING RFID" Kakatiya Institute of Technology and Science, Warangal
[18] You Only Look Once: Unified, Real-Time Object Detection
[20] Muhamad Ibrohim, Maya Selvia Lauryn,Rama Dhanan Jaya" RADIO FREQUENCY IDENTIFICATION (RFID) BASED EMPLOYEE ATTENDANCE SYSTEM DESIGN" Universitas Serang Raya
[21] Eliesalame” How to Connect a Serial LCD With an Arduino Nano”
[22] Nurkhasanah,Satria Maulana Ihsan,Rimulyo Wicaksono"PROTOTYPE OF WASTE COLLECTION TOOLS IN RIVERS BASED ON ANDROID"Universitas Negeri Jakarta