Visually, mango leaf plants monitoring system can increase optimal growth through early detection of leaf disease. However, the high complexity of mango leaves with multiple intersections and noisy backgrounds causes the model performance to be less sensitive to the actual condition of the leaf plant. To address this issue, the proposed model will be constructed by integrating YOLOv10 and the improved multi-scale attention module. Specifically, the detection head structure of YOLO will be fused with the attention module which consists of the integrated efficient multi-scale attention (EMA) module and non-local block (NLB) mechanism. This mechanism not only improves the model performance, especially in long-range dependencies and tiny objects, but also provides an adaptable lightweight model for edge computing systems. The experimental results indicate the proposed model has outstanding performance with mAP50 of 0.944. Meanwhile, the original model gains the mAP50 of 0.917. Compared with other models including YOLOv5s, YOLOv5s+CBAM, YOLOv8s, YOLOv8s+SAM, YOLOv9s and YOLOv10s, the proposed model achieves the best overall detection performance.

Agriculture is one of the sources of income a region can rely on to support its economy. Traditional agriculture relies primarily on human performance and observation, resulting in greater production costs and, subsequently, higher selling prices. Artificial intelligence-based technology can be used to reduce production costs, increase productivity, and provide consumer convenience. An indicator that is easy to interpret in measuring the quality and optimization of plant growth is the visualization of the condition of the leaves. The artificial intelligence technique that can be implemented in this regard is the object detection model. However, the challenge is the complex, multi-object, and multi-intersection condition of the leaves, which causes the model to be less optimal in conducting classification and detection tasks regarding whether the leaf condition is good or not. A YOLOv7 model will be employed in order to detect leaf quality, whether in an “optimal” or “not optimal” condition. To enhance the model's performance by improving accuracy through feature extraction enhancement, YOLOv7 will be integrated with the attention module, called the convolutional block attention module (CBAM). The case study in this research is detecting a mango plant which is one of the plants that can provide a high economic impact and the object observed is the mango plant leaf. Several previous studies related to the implementation of attention modules in object detection include the improved pest-YOLO for real-time pest detection by combining YOLOv3 with efficient channel attention (ECA) and a transformer encoder. The ECA module and transformer encoder were integrated into the backbone and neck block systems of YOLO [1]. The lightweight YOLO model combined with SE-CSPGhostnet by improving the backbone block which employs squeeze-and-excitation networks (SENet) and a convolution technique consisting of regular convolution and ghost convolution [2]. There is a highlighted improvement of YOLOv7 compared to the previous version of YOLO, which is Extended Efficient Layer Aggregation Networks (E-ELAN). YOLOv7's learning ability is enhanced by using this network while maintaining the transition layer's architecture. E-ELAN enhances the structure of computational blocks. E-ELAN can keep its transition layer while changing only the architectural component of the computational block. This allows the model to learn more different features. The fundamental purpose of model scaling is to alter certain model features in order to generate models at various scales to fulfill the needs of different inference speeds.

In this study, blockchain technology is used to strengthen the security of wireless sensing data in the network architecture of the Internet of Things. For many organic farms that contain crops with high unit prices, the environmental parameters of the crops grown on the farm are often regarded as confidential data. First, a wireless sensing network is used to transmit sensing data to the farm’s front-end integrated microcontroller for fusion of the sensing data. After fusion of the sensing data is completed, the processed data is transmitted to the data processing center of the entire farm for encapsulation of the blockchain algorithm. Data are sent to the cloud database for storage after they are packaged by the system. At the same time, the data in the cloud database can be managed and analyzed by remote operators. Data encapsulation based on blockchain technology is used to effectively prevent any data from being stolen or destroyed by hackers. A complete blockchain encapsulation database system is implemented in the experimental stage, and the operator can use this system to encrypt the data that have been fused at the remote end for blockchain technology. In the encryption process of this blockchain, the security of data processing can be specifically enhanced by the system. Finally, each encapsulated block of data is securely stored in a private cloud database. There is a function to check whether the data in the cloud have been tampered with. A very innovative approach to using blockchain technology is proposed in this article to enhance the security of data processing.

The goal of this research is to improve the wireless sensor network data transmission method to save energy consumption of sensor nodes and improve the overall network life cycle. This study uses the ant colony algorithm task allocation control strategy, the execution capability, energy consumption and lifespan of each central node to realize the real-time task allocation control. This study proposes to use an improved hybrid ant colony algorithm to find the shortest path of the cluster and transmit it to the confluence node, so as to avoid unnecessary energy consumption caused by long-distance transmission. Through the independent cooperation of each node and the completion of data processing, the analysis results are finally obtained through information fusion and comprehensive decision-making. The research experiments show that the peer-to-peer wireless network structure in this study not only reduces the burden of communication, calculation and energy consumption of a single central node, but also helps to improve the life and stability of the wireless sensing network.

This research project aims to build a LoRa-based Internet of Things (IoT) secure localization system and application based on multi-sensor fusion calculation. The system of this project comprises LoRa hosts, which receive the signals from various nodes, and are connected to a multi-sensor fusion arithmetic system through a wireless network. This study crosses domains and integrates related engineering automation, network security technology, multi-sensor fusion calculation design, and the LoRa localization technique, and the research findings are expected to contribute to the network security of the defense industry and research on the LoRa IoT localization system.

This paper proposes the adaptive functional coupling recurrent cerebellar model articulation controller (AFCRC). The AFCRC system contains an integrated error function, a TSK fuzzy compensator, and a novel cerebellar model articulation controller (CMAC), which is developed based on the concept of a recurrent neural networks (RNNs) and a functional coupling NN (FCNN). This study uses the proposed AFCRC to control the direct torque control drive system of a switched reluctance motor (SRM), and compares it with the traditional CMAC and FCMAC. The experimental results reveal that the root mean square error (RMSE) is used as a performance index for comparing of the traditional CMAC, FCMAC, and AFCRC, respectively. The results show that the proposed AFCRC exhibits the robustness against external disturbances. Thus, the proposed control strategy is advantageous at various speed commands and has improved dynamic responses.

This paper proposes an innovative approach to WSNs (wireless sensor networks) that establishes a mobile database. This mobile database uses embedded system hardware modules. This database is actually built into the embedded system. This advantage of this database is that it is very easy to collect information, very confidential and secure, and can instantly upload measured sensor data to the cloud-end. This wireless sensor network system has a relay station function. This proposed mobile database is also built on the same embedded hardware module, and can be connected to the private cloud data center. Sensors upload data to the database; the system then makes a visual comparison and draws the relevant historical data. Because our database is built onto an embedded operating system, our research uses the Python programming languages. This system uses measured sensor data to perform the modeling and visual steps. This system will then draw a specific graphic to facilitate the comparison and analysis for the researcher. This study makes the embedded system a mobile database for the current commercial embedded hardware module for mobile database performance comparison.

A novel adaptive sliding diagonal recurrent fuzzy cerebellar model articulation controller (ASDRC) is proposed by this study. ASDRC includes the inputs by a sliding surface into a diagonal recurrent fuzzy cerebellar model articulation controller (DRCMAC). The ASDRC enables cerebellar model articulation controller (CMAC) to exhibit both static and dynamic characteristics, indicating that the use of DRCMAC improves the disadvantage of conventional CMAC while exhibiting the advantages of fuzzy CMAC (FCMAC). Regarding the proposed ASDRC, the adaptive update law determining the memory weights, means of Gaussian functions, and standard deviations of Gaussian functions is yielded by the Lyapunov stability theory; moreover, the gradient descent method is applied to yield the recurrent weight update law. Using the adaptive update law and recurrent weight update law of the ASDRC to implement online adjustment ensures system stability. To demonstrate the performance of the proposed ASDRC, this study applies it to the direct torque control (DTC) systems of an induction motor to perform experiments. The root-mean-square error is used as an assessment indicator to compare the results of the proposed controller with those of the FCMAC. The experimental results prove that the ASDRC has more excellent response, and its performance is superior to that of the FCMAC.

This paper presents a framework for hand gesture recognition based on 3-channel electromyography (EMG) sensors. In the framework, the start and the end points of meaningful gesture segments are detected automatically by checking the cross points of EMG signals and their moving average curves. Then, a classifier combining k-Nearest Neighbor (kNN) and Decision Tree algorithms is used for achieving gesture recognition. For gesture-based control application, a real-time interactive system has been built up for household appliance using 10 kinds of hand gestures as control commands. Our proposed framework facilitates intelligent and natural control in gesture-based interaction.

This study builds an automatic monitoring system for the fish pond culture environment. The purpose of this study is to reduce the financial losses in fish culture stock resulting from natural disasters, and facilitate better control of the fish pond environment. The physical water quality signals are therefore extracted using temperature, dissolved oxygen and pH sensing modules. The water heater, submerged motor pump, air pump, feeding trough and LED illuminating lamp are controlled to improve the water quality and reduce labor. This study utilizes the self-adaptive data fusion approach via a wireless sensors network framework to remotely monitor the fish pond automatic control system. The experimental results show that the fish pond culture environment can be accurately and stably monitored.

A number of ZigBee based monitoring systems are built on the basis of IOT technology in this work on the perception layer, using temperature/humidity sensors, light sensors and 3-axis accelerometer modules. Wirelessly transmitted to a monitoring center, all the sensed data are collected by a human computer interface. On the application layer in an IOT, simulation experiments are conducted, namely, applications of light sensors to an automated basketball court lighting system, 3-axis accelerometer modules to the monitoring of infant's sleeping posture and accidental fall of the elderly, and temperature/humidity sensors to thermal comfort testing. This research work is validated as an effective way to achieve the aim of power consumption reduction, improve the health care quality and provide a higher comfort level.

This panel will establish an automated monitoring system of wireless sensor networks for a fish farm Environment Simulation. This system allows a user with a mobile device to monitor the fish farm Environmental Data with Instant mastery and control over the various environmental data. Temperature, dissolved oxygen, PH value and water level sensing modules are incorporated in this monitoring system. MCU processing is used to capture the physical sensing signal. The ZigBee wireless sensor network brings the data to a central processing core. A WIFI interface transfers the data to the user terminal device. The user can control the entire fish farm environment through the terminal device. Android software was used to design the terminal device user interface. A low power MSP430 series MCU is the core of each sensing terminal and the central terminal. The power supply can be battery-powered, standard electricity supply and/or solar battery powered. UPS makes the whole system more secure with low-cost, low energy consumption, easy operating features with a high degree of freedom for this wireless breeding environment monitoring system.

This study proposed an IoT (Internet of Things) system for the monitoring and control of the aquaculture platform. The proposed system is network surveillance combined with mobile devices and a remote platform to collect real-time farm environmental information. The real-time data is captured and displayed via ZigBee wireless transmission signal transmitter to remote computer terminals. This study permits real-time observation and control of aquaculture platform with dissolved oxygen sensors, temperature sensing elements using A/D and microcontrollers signal conversion. The proposed system will use municipal electricity coupled with a battery power source to provide power with battery intervention if municipal power is interrupted. This study is to make the best fusion value of multi-odometer measurement data for optimization via the maximum likelihood estimation (MLE).Finally, this paper have good efficient and precise computing in the experimental results.

This paper proposed environmental parameters are collected by use of outdoor ZigBee based weather stations as a prerequisite for the optimisation of plant growth. In most cases, all the sensors required are integrated into a weather station, due to which merely a single monitoring node is employed following data aggregation. An energy efficient center data aggregation algorithm, where an ant colony algorithm is applied to the construction of a level gradient field, is presented as an effective way to extend the life cycles of sensor nodes. A weather station and a ZigBee module both are portable and easy to install battery operated devices. Furthermore, a remote web‐based human machine interface (HMI) is developed by InduSof on a server, and has an access to a database. This proposed algorithm is confirmed by computer simulations as an effective approach to remarkably extend the life cycles of sensor nodes. This work can be applied not merely to traditional outdoor large scale farming, but also to small scale indoor plantation, e.g. in a green house, a plant factory, etc., and applied to the field of conservation ecology.

This work builds an intelligent environmental monitoring system for industrial production and safety concerns. There are a number of smart monitoring systems developed for smart grids, bridges or machine system management, temperature/humidity monitoring and so on. As information technology advances at a continuous rapid pace, humans have become connected online and this connection has extended into the interactions between things and humans. Employing the FLAG‐PSoC‐1605A development board as the platform, all sensor data are transmitted via a ZigBee wireless module, a TCP/IP network module and a long range wireless communication GPRS/SMS module to a remote end PC for analysis. This constitutes a smart network for connections between humans and objects.

Accelerating a protein folding by implementing it in a reconfigurable field programmable gate array (FPGA) is described. This paper presents a methodology for the design of a reconfigurable computing system applied to a complex problem in molecular biology: the protein folding problem. This paper employed VMD tool and force field simulation theorem based on FPGA for protein folding solution. This technique consists of two components: finished protein folding process and found out active sites for drug docking. The goal of protein folding simulation is to search the global energy minimum location with stability state and the when the protein is finished the folding task, we can find out the active sites for pre-process of ligand protein docking. An efficient hardware-based approach was devised to achieve a significant reduction of the search space of possible foldings. Several simulations were done to evaluate the performance of the system as well as the demand for FPGApsilas resources.

This study attempts to apply pattern recognition (PR) technologies with associative memory to real-time pattern recognition of engineering components using a client-server network structure in a Web-based recognition system. A remote engineer is able to draw directly the shape of engineering components using the browser, and the recognition system will search for the component database of a company using the Internet. Component patterns are stored in the database system. Their properties and specifications are also attached to the data field of each component pattern except that of engineering components. In our approach, the recognition system adopts parallel computing, and will raise the recognition rate. Our recognition system is a client-server network structure using the Internet. The system uses a recurrent neural network (RNN) with associative memory to perform training and recognition. The last phase utilizes the technology of database matching and solves the problem of spurious state. Our system will be used at the Yang-Fen Automation Electrical Engineering Company.

The aim of the paper is to propose a three-dimensional graphics chip with parallel architecture, in accordance with the chip of the NURBS algorithm was structured. This architecture presents a regular and easily scalable structure, suitable for VLSI implementation, which can be efficiently exploited for the computing process. This architecture can apply to a peripheral device of a computer and a user can use it for fitting curves and surfaces, convert NURBS curves to Bezier and so on, which have 16 bit precision. In this paper, we first describe the framework of the whole system. Secondly, we introduce the hardware of FPGA and explain the principle. Next, we will also illustrate NURBS, and test and verify using Visual C++ and OpenGL. The performance of the proposed architecture is improved by the use of carry save arithmetic which permits the reduction of the system time cycle.

This study attempts to apply the principle of neural networks and pattern recognition (PR) technologies to real-time recognition by client-server network structure into a web-based recognizing system. In this paper, we recommend a Web-Based PR technology, which is improved recurrent neural network (RNN) from possessing feedback and non-linear activation function with its input, be taken out threshold. The purpose of this article is. to construct a Client-Server network structure for PR system with associative memory. The Server-end is built a databases management system for storage sample patterns. In proceed with training, the user can real-time assign any pattern, which is a record in the Server-end databases. In deal with retrieve task, we propose a novel PR method via databases matching, it can efficient solve spurious states problem from RNN in the WBPR system. On the other hand, taking advantage of Database Matching is to overcome the capacity restrictions on RNN. In order to clarify and corroborate the above Web-Based PR technology, thus a simulation experiment will be presented and their algorithms are also discussed.

VR technologies provide a unique method for enhancing user visualization of complex three-dimensional graphics and environments. The study attempts to apply virtual reality (VR) technologies to a computer-aided design (CAD) curriculum by integration of network, CAD and VR on a Web-based learning environment. Through VR technologies, it is expected that the traditional two-dimensional computer graphics (CG) course can be expanded into a three-dimensional real-time simulation CG course. The use of these modern education technologies can be used to improve the effectiveness of learning system dynamics in CG courses. We present how these educational issues can assist and improve with the development of a Web-based learning system, we call WebDeGrator (Web-based Interactive Design Graphics). The design of a WebDeGrator learning system and the related themes of educational and VR technologies are presented and discussed. Future developments of the proposed Web-based learning framework are also discussed.