The real-time target detection of crop pests can help detect and control pests in time. In this study, we built a lightweight agricultural pest identification method based on modified Yolov5s and reconstructed the original backbone network in tandem with MobileNetV3 to considerably reduce the number of parameters in the network model. At the same time, the ECA attention mechanism was introduced into the MobileNetV3 shallow network to meet the aim of effectively enhancing the network’s performance by introducing a limited number of parameters. A weighted bidirectional feature pyramid network (BiFPN) was utilized to replace the path aggregation network (PAnet) in the neck network to boost the feature extraction of tiny targets. The SIoU loss function was utilized to replace the CIoU loss function to increase the convergence speed and accuracy of the model prediction frame. The updated model was designated ECMB-Yolov5. In this study, we conducted experiments on eight types of common pest dataset photos, and comparative experiments were conducted using common target identification methods. The final model was implemented on an embedded device, the Jetson Nano, for real-time detection, which gave a reference for further application to UAV or unmanned cart real-time detection systems. The experimental results indicated that ECMB-Yolov5 decreased the number of parameters by 80.3% and mAP by 0.8% compared to the Yolov5s model. The real-time detection speed deployed on embedded devices reached 15.2 FPS, which was 5.7 FPS higher than the original model. mAP was improved by 7.1%, 7.3%, 9.9%, and 8.4% for ECMB-Yolov5 compared to Faster R-CNN, Yolov3, Yolov4, and Yolov4-tiny models, respectively. It was verified through experiments that the improved lightweight method in this study had a high detection accuracy while significantly reducing the number of parameters and accomplishing real-time detection.

This paper reports on investigations of the air-coupled ultrasonic (ACU) method to detect common defects in solid timber panels made of Chinese fir (Cunninghamia lanceolata (Lamb.) Hook.). The ACU technology is a non-contact method for nondestructive timber testing with quicker scanning rates compared to contact methods. A testbed was set up consisting of commercially available piezo-ceramic ACU transducers and in-house manufactured signal processing circuits. To demonstrate the suitability of the ACU technique, through-transmission measurement results are presented for samples with defects such as knots, wormholes, and cracks. Pulse compression methods (Barker-coded method) were used to improve the power of received signals based on cross-correction algorithms. Results showed defects of timber panels made of Chinese fir can be detected with a thickness of less than 40 mm. Defects larger than 3 mm in diameter could be detected with high precision. Applying the pulse compression method showed better results than using common sine signals as excitation signals since it increased the signal-to-noise ratio, which is especially important for air-coupled measurement of high-attenuation materials like timber materials. The measurement results on reference samples demonstrated that ACU technology is a promising method for timber defect detection, especially for the quality assessment of engineered wood products.

To overcome the problems of the low signal-to-noise ratio and poor performance of wood ultrasonic images caused by ring-down vibrations during the ultrasonic quality detection of wood, a composite pulse excitation technique using a wood air-coupled ultrasonic detection system is proposed. Through a mathematical analysis of the output of the ultrasonic transducer, the conditions necessary for implementing composite pulse excitation were analyzed and established, and its feasibility was verified through COMSOL simulations. Firstly, wood samples with knot and pit defects were used as experimental samples. We refined the parameters for the composite pulse excitation technique by conducting A-scan measurements on both defective and non-defective areas of the samples. Moreover, two stepper motors were employed to control the path for C-scan imaging to detect wood defects. The experiment results showed that the composite pulse excitation technique significantly enhanced the precision of nondestructive ultrasonic testing for wood defects compared to the traditional single-pulse excitation method. This technique successfully achieved precise detection and location of pit defects, with a detection accuracy rate of 90% for knot defects.

Air-coupled ultrasound overcomes the limitations of traditional contact-based ultrasonic methods that rely on liquid couplants. Still, it faces challenges due to the acoustic impedance mismatch between air and wood, causing significant signal scattering and attenuation. This results in weak transmission signals contaminated by clutter and noise, compromising measurement accuracy. This study proposes a coded pulse air-coupled ultrasonic method for detecting defects in wood. The method utilizes Golay code complementary sequences (GCCSs) to generate excitation signals, with its feasibility validated through mathematical analysis and simulations. A-scan imaging was performed to analyze the differences in signal characteristics between defective and non-defective areas, while C-scan imaging facilitated a quantitative assessment of defects. Experimental results demonstrated that GCCS-enhanced signals improved the ultrasonic penetration and axial resolution compared to conventional multi-pulse excitation. The method effectively identified defects such as knots and pits, achieving a coincidence area of 85% and significantly enhancing the detection accuracy.

The moving-magnet linear motor has received considerable attention in the development of logistic and factory automation in recent years. A reliable position detection system is the key to achieving the precise position and control of the motor. At present, the magnetic grid-scale and grating-scale are the most widely used traditional detection methods. However, these are not suitable for position detection with moving-magnet linear motors. They have the disadvantages of being easy to disturb, having a high cost, and exhibiting a limited measurement range. In this work, a moving-magnet linear motor position detection system based on an array of magnetoresistive sensors is used. The array is configured by arranging the magnetoresistive sensors at equal intervals along a line parallel to the trajectory of the armature. Then, the permanent magnet is fixed on the rotor and detected by sensors. When the rotor crosses the sensors in a parallel line, the changes in the magnetic field cause the magnetoresistive sensors to output two voltage signals directly proportional to the corresponding position changes. The signals are collected by the AD7606 and transmitted to the FPGA and STM32 controller for data processing, and the actual position of the rotor is calculated. This method has no length limitation and can be used for long-distance position detection. The experimental results show that the position detection system has a higher linear correlation coefficient compared with the magnetic grid ruler, in addition to a capability of ±9 μm accuracy, which verifies the validity of the position detection method for the moving-magnet linear motor.