Search

Abstract Background The lint percentage of seed cotton is one of the most important parameters for evaluating seed cotton quality and affects its price. The traditional measuring method of lint percentage is labor-intensive and time-consuming; thus, an efficient and accurate measurement method is needed. In recent years, classification-based deep learning and computer vision have shown promise in solving various classification tasks. Results In this study, we propose a new approach for detecting the lint percentage using MobileNetV2 and transfer learning. The model is deployed on a lint percentage detection instrument, which can rapidly and accurately determine the lint percentage of seed cotton. We evaluated the performance of the proposed approach using a dataset comprising 66 924 seed cotton images from different regions of China. The results of the experiments showed that the model with transfer learning achieved an average classification accuracy of 98.43%, with an average precision of 94.97%, an average recall of 95.26%, and an average F1-score of 95.20%. Furthermore, the proposed classification model achieved an average accuracy of 97.22% in calculating the lint percentage, showing no significant difference from the performance of experts (independent-sample t-test, t = 0.019, P = 0.860). Conclusion This study demonstrated the effectiveness of the MobileNetV2 model and transfer learning in calculating the lint percentage of seed cotton. The proposed approach is a promising alternative to traditional methods, providing a rapid and accurate solution for the industry.

Lint percentage of seed cotton is one of the important bases for pricing in the trading segment. Unfortunately, the conventional methods of lint percentage are manually operated, which relies on the abundant experience of experts, and restrained by personal, physical and environmental factors. Up to date, the calculation of the lint percentage of seed cotton has not fully automated. In this paper, we proposed a non-destructive detection method for automatically obtaining lint percentage of seed cotton based on optical penetration imaging and machine vision, for the first time to our knowledge. The cotton seed image was obtained by the penetration imaging setup with a LED white backlight source. To accurately identify the number of cotton seeds, the image features of the cotton seed was studied and three key features was been found, which are the circumference, area, and greyscale value, respectively. A calculation system based on the three key features was presented to process the images and then automatically calculate the lint percentage of seed cotton. The first step of the system is to segment the original image using adaptive thresholding followed by morphological operations. Afterwards, the number of cotton seed was obtained by the three key features of the cotton seed. Then, the lint percentage was achieved by a professional industry formula. The suggested lint percentage detection methods were verified by the experiments with two seed cotton varieties samples of H219 and ZHM19. The experimental results indicated that the detection average accuracy of the developed system for seed cotton varieties H219 and ZHM19 were 96.33% and 95.40%. [ABSTRACT FROM AUTHOR]

Bionic joints are crucial for robotic motion and are a hot topic in robotics research. Among various actuators for joints, shape memory alloys (SMAs) have attracted significant interest due to their similarity to natural muscles. SMA exhibits the shape memory effect (SME) based on martensite-to-austenite transformation and its inverse, which allows for force and displacement output through low-voltage heating. However, one of the main challenges with SMA is its limited axial stroke. In this article, a bionic joint based on SMA wires and a differential pulley set structure was proposed. The axial stroke of the SMA wires was converted into rotational motion by the stroke amplification of the differential pulley set, enabling the joint to rotate by a sufficient angle. We modeled the bionic joint and designed a proportional–integral (PI) controller. We demonstrated that the bionic joint exhibited good position control performance, achieving a rotation angle range of −30° to 30°. The proposed bionic joint, utilizing SMA wires and a differential pulley set, offers an innovative solution for enhancing the range of motion in SMA actuated bionic joints.

Path planning is a crucial concern in the field of mobile robotics, particularly in complex scenarios featuring narrow passages. Sampling-based planners, such as the widely utilized probabilistic roadmap (PRM), have been extensively employed in various robot applications. However, PRM’s utilization of random node sampling often results in disconnected graphs, posing a significant challenge when dealing with narrow passages. In order to tackle this issue, we present equipotential line sampling strategy for probabilistic roadmap (EPL-PRM), a novel approach derived from PRM. This paper initially proposes a sampling potential field, followed by the construction of equipotential lines that are denser in the proximity of obstacles and narrow passages. Random sampling is subsequently conducted along these lines. Consequently, the sampling strategy enhances the likelihood of sampling nodes around obstacles and narrow passages, thereby addressing the issue of sparsity encountered in traditional sampling-based planners. Furthermore, we introduce a nodal optimization method based on an artificial repulsive field, which prompts sampled nodes to move in the direction of repulsion. As a result, nodes around obstacles are distributed more uniformly, while nodes within narrow passages gravitate toward the middle of the passages. Finally, extensive simulations are conducted to evaluate the proposed method. The results demonstrate that our approach achieves path planning with superior efficiency, lower cost, and higher reliability compared with traditional algorithms.

Inspired by the natural skeletal muscles, this paper presents a novel shape memory alloy-based artificial muscle matrix (AMM) with advantages of a large output force and displacement, flexibility, and compactness. According to the composition of the AMM, we propose a matrix control strategy to achieve independent control of the output force and displacement of the AMM. Based on the kinematics simulation and experiments, we obtained the output displacement and bearing capacity of the smart digital structure (SDS) and confirmed the effectiveness of the matrix control strategy to achieve force and displacement output independently and controllably. A bionic mechanical ankle actuated by AMM was proposed to demonstrate the actuating capability of the AMM. Experimental results show that the angle and force of the bionic mechanical ankle are output independently and have a significant gradient. In addition, by using a self-sensing method (resistance self-feedback) and PD control strategy, the output angle and force of the bionic mechanical ankle can be maintained for a long time without overheating of the AMM.

Existing thruster fault diagnosis methods for AUV (autonomous underwater vehicle) usually need sufficient labeled training data. However, it is unrealistic to get sufficient labeled training data for each working condition in practice. Based on this challenge, a transferable thruster fault diagnosis approach is proposed. In the approach, an IPSE (instantaneous power spectrum entropy) and a STNED (signal-to-noise energy difference) are added to SPWVD (smoothed pseudo Wigner-Ville distribution) to identify time and frequency boundaries of the local region in the time-frequency power spectrum caused by thruster fault, forming a TFE (time-frequency energy) method for feature extraction. In addition, the RCQFFV (relative change quantity of the fault feature value), an MSN (multiple scale normalization) and a LSP (least square prediction) are added to SVDD (support vector data description) to align distributions of fault samples, contributing a TSVDD (transferable SVDD) for classification of fault samples. The experimental results of a prototype AUV indicate that the fault feature is monotonic to the percentage of thrust loss for the proposed TFE but not for the SPWVD. The TSVDD has a higher overall classification accuracy in comparison to conventional SVDD under working conditions with no labeled training data.

Amino acid decarboxylases convert amino acids into different biogenic amines which regulate diverse biological processes. Therefore, identifying the substrates of amino acid decarboxylases is critical for investigating the function of the decarboxylases, especially for the new genes predicted to be amino acid decarboxylases. In the present work, we have established a simple and efficient method to identify the substrates and enzymatic activity of amino acid decarboxylases based on LC-MS methods. We chose GAD65 and AADC as models to validate our method. GAD65 and AADC were expressed in HEK 293T cells and purified through immunoprecipitation. The purified amino acid decarboxylases were subjected to enzymatic reaction with different substrate mixtures in vitro. LC-MS analysis of the reaction mixture identified depleted or accumulated metabolites, which corresponded to candidate enzyme substrates and products, respectively. Our method successfully identified the substrates and products of known amino acid decarboxylases. In summary, our method can efficiently identify the substrates and products of amino acid decarboxylases, which will facilitate future amino acid decarboxylase studies.