Biosystems Engineering最新文献

{"title":"Spatial LiDAR odometry and mapping for complex agricultural environments - Spatial FieldLOAM","authors":"Jurij Rakun ,&nbsp;František Duchoň ,&nbsp;Peter Lepej","doi":"10.1016/j.biosystemseng.2024.09.020","DOIUrl":"10.1016/j.biosystemseng.2024.09.020","url":null,"abstract":"<div><div>The challenge of autonomous driving in natural environments, without the use of GNSS devices is addressed. It utilises the readings from a multichannel LiDAR, supported by IMU, and enhances the capabilities of the FieldSLAM algorithm to establish an independent localisation and mapping system. This system is designed for performing specific tasks in predefined agricultural areas, employing incremental LOAM techniques. By comparing the outcomes of the novel Spatial FieldLOAM algorithm with the assistance of a precise Inertial Measurement Unit (IMU) and using the state-of-the-art RTK-GPS system as the ground truth, it is concluded that the Spatial FieldLOAM achieves an error rate of 5.5%, whereas the Xsens IMU yields an error rate of 5.7%. In terms of Euclidean distances to the final RTK GPS supported localisation on a 68.7 m test run, the error rates are 3.78 m and 3.92 m, respectively, or 0.0038 m per epoch for the Spatial FieldLOAM algorithm during non-vegetation season. The tests were also conducted during the vegetation season in a total length of 210 m, revealing a difference of 3.07 m distance between the final position calculated by the Spatial FieldLOAM and Xsens IMU.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"248 ","pages":"Pages 58-72"},"PeriodicalIF":4.4,"publicationDate":"2024-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142421060","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating a hybrid process of anaerobic digestion, aerobic degradation, and electrochemical separation for swine wastewater treatment with methane and nutrient recovery","authors":"Run-Feng Chen ,&nbsp;Chun-Hai Wei ,&nbsp;Hai-Tao Zhong ,&nbsp;Xiu-Feng Ye ,&nbsp;Jun-Jie Ye ,&nbsp;Kai Liu ,&nbsp;Quan-Bao Zhao ,&nbsp;Huu Hao Ngo","doi":"10.1016/j.biosystemseng.2024.09.022","DOIUrl":"10.1016/j.biosystemseng.2024.09.022","url":null,"abstract":"<div><div>A hybrid process of anaerobic digestion (AD), aerobic degradation, and electrochemical separation was evaluated for treating real swine wastewater that is rich in organic and nutrient to achieve methane and nutrient recovery and industry standard discharge quality. Fe anode electrocoagulation and Mg anode struvite electrochemical precipitation (SEP) were evaluated as AD pretreatments. Both removed partial chemical oxygen demand (COD) from raw swine wastewater, but only SEP slightly enhanced the methane yield of pretreated swine wastewater. The SEP efficiency of the AD effluent was significantly better than raw swine wastewater. A further coupled micro/ultra-filtration produced high-purity (96%) struvite. SEP and struvite chemical precipitation (SCP) were evaluated for AD effluent treatment. This showed that compared with SCP following first-order reaction kinetics (reaction rate constant of 0.791 and 0.854 h⁻¹ for NH₄⁺-N and PO₄^3--P), SEP not only achieved better removal of COD, NH₄⁺-N and PO₄^3--P, but was also shown to follow zero-order reaction kinetics (reaction rate constant of 5.72 and 5.78 mmol L⁻¹ h⁻¹ for NH₄⁺-N and PO₄^3--P). The SEP and SCP treated AD effluent was evaluated by conventional activated sludge (CAS), showing faster COD removal (first-order reaction rate constant of 0. 213 and 0.163 h⁻¹) and lower residual COD (150 and 248 mg L⁻¹) from SEP than SCP treated AD effluent, making the final effluent well below Chinese livestock wastewater discharge standards. Therefore, an emerging hybrid anaerobic membrane bioreactor (AnMBR)-SEP-CAS is proposed for swine wastewater treatment and proved to be more economically viable than the conventional hybrid AD-SCP-CAS process via cost-benefit analysis.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"248 ","pages":"Pages 47-57"},"PeriodicalIF":4.4,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142421059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PointStack based 3D automatic body measurement for goat phenotypic information acquisition","authors":"Bo Jin ,&nbsp;Guorui Wang ,&nbsp;Jingze Feng ,&nbsp;Yongliang Qiao ,&nbsp;Zhifeng Yao ,&nbsp;Mei Li ,&nbsp;Meili Wang","doi":"10.1016/j.biosystemseng.2024.09.008","DOIUrl":"10.1016/j.biosystemseng.2024.09.008","url":null,"abstract":"<div><div>The body size of livestock is an essential phenotypic trait in genetic breeding, gene improvement, health screening, and animal welfare. To develop a non-contact automatic system for measuring goat body traits, we propose a point-cloud segmentation model based on an improved PointStack, which segments the automatically acquired three-dimensional (3D) point-cloud data of goats into different parts, including the head, front legs, hind legs, chest, abdomen, hip, and tail. The segmented point cloud, along with the physiological features of the goat, is then used to locate the corresponding key points for body size measurement. A novel method for key point localisation is proposed that includes coordinate normalisation, retrieval of key clusters, key point adjustment, optimisation of the traveling salesman problem, and edge detection. These methods were designed to reduce discrepancies at crucial points of body features, thereby facilitating the precise computation of the body size parameter in goats. In this work, 326 point clouds representing the upright posture of 55 goats were used for segmentation and body size measurement testing. The proposed segmentation model achieved a mean intersection over union of 89.21% and accuracy of 94.54%, outperforming comparative models. In the body traits measurement experiment, mean absolute percentage errors for body length, body height, chest width, chest girth, hip height, and hip width were recorded as 3.24%, 2.54%, 5.43%, 3.08%, 2.16%, and 4.59%, respectively. In summary, the proposed automated measurement method demonstrates high accuracy, strong robustness, and holds significant potential for widespread application.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"248 ","pages":"Pages 32-46"},"PeriodicalIF":4.4,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142357508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Semantics-aware next-best-view planning for efficient search and detection of task-relevant plant parts","authors":"Akshay K. Burusa,&nbsp;Joost Scholten,&nbsp;Xin Wang,&nbsp;David Rapado-Rincón,&nbsp;Eldert J. van Henten,&nbsp;Gert Kootstra","doi":"10.1016/j.biosystemseng.2024.09.018","DOIUrl":"10.1016/j.biosystemseng.2024.09.018","url":null,"abstract":"<div><div>Searching and detecting the task-relevant parts of plants is important to automate harvesting and de-leafing of tomato plants using robots. This is challenging due to high levels of occlusion in tomato plants. Active vision is a promising approach in which the robot strategically plans its camera viewpoints to overcome occlusion and improve perception accuracy. However, current active-vision algorithms cannot differentiate between relevant and irrelevant plant parts and spend time on perceiving irrelevant plant parts. This work proposed a semantics-aware active-vision strategy that uses semantic information to identify the relevant plant parts and prioritise them during view planning. The proposed strategy was evaluated on the task of searching and detecting the relevant plant parts using simulation and real-world experiments. In simulation experiments, the semantics-aware strategy proposed could search and detect 81.8% of the relevant plant parts using nine viewpoints. It was significantly faster and detected more plant parts than predefined, random, and volumetric active-vision strategies that do not use semantic information. The strategy proposed was also robust to uncertainty in plant and plant-part positions, plant complexity, and different viewpoint-sampling strategies. In real-world experiments, the semantics-aware strategy could search and detect 82.7% of the relevant plant parts using seven viewpoints, under complex greenhouse conditions with natural variation and occlusion, natural illumination, sensor noise, and uncertainty in camera poses. The results of this work clearly indicate the advantage of using semantics-aware active vision for targeted perception of plant parts and its applicability in the real world. It can significantly improve the efficiency of automated harvesting and de-leafing in tomato crop production.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"248 ","pages":"Pages 1-14"},"PeriodicalIF":4.4,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142357509","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting wheat scab levels based on rotation detector and Swin classifier","authors":"Dongyan Zhang ,&nbsp;Zhipeng Chen ,&nbsp;Hansen Luo ,&nbsp;Gensheng Hu ,&nbsp;Xin-Gen Zhou ,&nbsp;Chunyan Gu ,&nbsp;Liping Li ,&nbsp;Wei Guo","doi":"10.1016/j.biosystemseng.2024.09.016","DOIUrl":"10.1016/j.biosystemseng.2024.09.016","url":null,"abstract":"<div><div>Wheat scab is a highly destructive disease that adversely impact wheat crops throughout their growth cycle. It is crucial to promptly evaluate the levels of wheat scab in the field to prevent its spread. Manual observation, however, is inefficient and time-consuming. Recent research has indicated that computer vision-based methods can enhance efficiency in this regard. This study proposed a method for predicting wheat scab levels using a rotation detector and Swin classifier.</div><div>To minimise background interference, the study incorporated the rotation wheat detector (RWD) network for detecting wheat heads. The RWD network employed the Kalman filter Intersection over Union (KFIoU) to predict the angle, thereby improving accuracy. The Swin wheat classifier (SWC) network was employed to classify healthy and diseased wheat heads. The SWC network benefited from the shifted window self-attention module (SW-MSA), which enhanced feature extraction by establishing connections with other windows. The proposed method was evaluated using wheat field images collected over 3 years. The results demonstrate promising performance, achieving a 96% accuracy in predicting wheat scab levels. Furthermore, the <em>R</em>^<em>2</em> and RMSE values for diseased wheat count were 97.62% and 3.61, respectively. This method offers an accurate means of predicting wheat scab levels through the analysis of wheat field images. Additionally, the introduction of the rotation detector presents a novel contribution to research on wheat scab detection.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"248 ","pages":"Pages 15-31"},"PeriodicalIF":4.4,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142357510","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impacts of crate design, number of heat lamps and lying posture on the occurrence of shoulder lesions in sows","authors":"Shubham Bery ,&nbsp;Tami M. Brown-Brandl ,&nbsp;Gary A. Rohrer ,&nbsp;Sudhendu Raj Sharma ,&nbsp;Suzanne M. Leonard","doi":"10.1016/j.biosystemseng.2024.09.017","DOIUrl":"10.1016/j.biosystemseng.2024.09.017","url":null,"abstract":"<div><div>This study investigated the interaction of sow and engineering factors on shoulder lesion formation. Sows were randomly assigned to three farrowing crate designs: Traditional Stall Layout, Expanded Creep Stall Layout, and Expanded Sow &amp; Creep Stall Layout. Each crate configuration was further differentiated by the inclusion of either one (1HL) or two (2HL) heat lamps. Digital and depth images were collected from an overhead time of flight depth camera (Kinect V2) every 5 s. Computer vision techniques were employed to analyze top-down digital images from the 21st to the 24th day of farrowing to detect and estimate lesion size. Additionally, the study incorporated an analysis of sow lying behaviors on the occurrence and size of lesions using depth images. Sow's environmental and phenotypic data - weight, parity, body condition score, total lying time and number of lying transitions in a day were investigated for impact on shoulder lesion. The results indicated that the interaction of smaller crate sizes and increased heat lamp usage significantly impacted lesion occurrence (p &lt; 0.05). Also, higher parity and lighter weight sows showed higher lesion occurrence (p &lt; 0.05). However, other factors, such as the number of heat lamps alone and detailed metrics of lying postures, did not show a significant impact on lesion occurrence. In contrast, none of the studied factors showed a significant impact on the size of shoulder lesions. This highlights the importance of allocating crate space with respect to heat lamp placement to the sows.</div><div>Science4Impact Statement (S4IS): This manuscript evaluates shoulder lesions' presence and size in lactating sows housed within farrowing stalls. Shoulder lesions are one of the main causes of premature culling in sows and are a major concern for animal well-being. Understanding the impact of crate design and the number of heat lamps is important for the engineering design of the farrowing environment.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"247 ","pages":"Pages 249-256"},"PeriodicalIF":4.4,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142314111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Characterising equivalent droplet indicators of sprinkler irrigation from a kinetic energy perspective","authors":"Rui Zhang ,&nbsp;Yichuan Liu ,&nbsp;Delan Zhu ,&nbsp;Pute Wu ,&nbsp;Changjuan Zheng ,&nbsp;Xiaomin Zhang ,&nbsp;Nazarov Khudayberdi ,&nbsp;Changxin Liu","doi":"10.1016/j.biosystemseng.2024.09.019","DOIUrl":"10.1016/j.biosystemseng.2024.09.019","url":null,"abstract":"<div><div>Equivalent droplet velocity and diameter are important parameters for measuring the effectiveness of sprinkler spraying; however, non-optical test methods (paper stain, flour pellet, and oil immersion methods) can only obtain the droplet number and diameter. With the widespread use of optical instruments in sprinkler testing, droplet velocity can also be measured, therefore, it has become possible to calculate the average droplet characteristics from an energy perspective. This paper proposes an energy-weighted method for calculating droplet equivalence indicators. Statistical analyses were performed based on five types of sprinkler irrigation droplet distribution data to compare the characteristics and differences between the energy-weighted method and the calculation results of the other methods. The results showed that 1) the velocity outcomes of the energy-weighted droplet equivalent method, empirical formula I, and empirical formula II consistently increase and decrease; 2) the equivalent droplet diameter based on the energy-weighted method is the largest, followed by the equivalent method related to droplet volume, and the smallest is the equivalent method related to droplet quantity; and 3) the equivalent droplet velocity and diameter calculated by the energy-weighted equivalent method can characterise droplets with a high energy contribution. The energy-weighted equivalent droplet velocity and diameter indicators derived in this study provide new ideas for characterising droplet averaging.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"247 ","pages":"Pages 241-248"},"PeriodicalIF":4.4,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142314112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Disturbance analysis and seeding performance evaluation of a pneumatic-seed spoon interactive precision maize seed-metering device for plot planting","authors":"Shidong Deng,&nbsp;Yamei Feng,&nbsp;Xiupei Cheng,&nbsp;Xianliang Wang,&nbsp;Xiangcai Zhang,&nbsp;Zhongcai Wei","doi":"10.1016/j.biosystemseng.2024.09.007","DOIUrl":"10.1016/j.biosystemseng.2024.09.007","url":null,"abstract":"<div><div>In response to the serious issue of missed seeding of the seed-metering device caused by the small and gradually decreasing number of maize seeds in plot planting conditions, a precision seed-metering device for maize plots with pneumatic-seed spoon interactive was designed. The seed-metering device utilises the coupling interaction between the seed spoons and airflow to adjust the maize posture in the filling zone, achieving stable filling of the seed-metering device with a low population of seeds in the seed chamber. EDEM software is used to simulate and analyse the disturbance caused by three types of seed-metering discs and the average kinetic energy in the filling zone as the evaluation criterion. A test platform for seed-metering device of maize plot was constructed, with the qualified index, multiple index, and missing index as evaluation criteria. A full-factorial experiment was conducted with rotation speed of the seed-metering disc, air pressure, and types of seed-metering discs as factors, determining the optimal seed-metering disc for seeding performance. The results indicated that under conditions of low seed population in the seeding chamber, with air pressures ranging from −1.5 to −2.5 kPa and seed-metering disc speeds between 1.16 and 3.49 rad s⁻¹, the seed-metering device with linear disturbance exhibited a multiple index of &lt;6.25% and a missing index of &lt;3.46%. Additionally, the qualified index consistently reached 90.29%. These evaluation criteria meet the standards, demonstrating effective seeding capabilities.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"247 ","pages":"Pages 221-240"},"PeriodicalIF":4.4,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142311072","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Positioning of mango picking point using an improved YOLOv8 architecture with object detection and instance segmentation","authors":"Hongwei Li ,&nbsp;Jianzhi Huang ,&nbsp;Zenan Gu ,&nbsp;Deqiang He ,&nbsp;Junduan Huang ,&nbsp;Chenglin Wang","doi":"10.1016/j.biosystemseng.2024.09.015","DOIUrl":"10.1016/j.biosystemseng.2024.09.015","url":null,"abstract":"<div><div>Positioning of mango picking points is a crucial technology for the realisation of automated robotic mango harvesting. Herein, this study reported a visualised end-to-end system for mango picking point positioning using improved YOLOv8 architecture with object detection and instance segmentation, as well as an algorithm of picking point positioning. At first, the improved YOLOv8n model, incorporating the BiFPN structure and the SPD-Conv module, was utilised to enhance the detection performance of mango fruits and stems. This model achieved a detection precision of 98.9% in fruits and 97.1% in stems, with recall of 99.5% and 94.6% respectively. Then, the YOLOv8n-seg model was used for segment the stem ROI (Region of interest), leading to 81.85% in MIoU and 88.69% in mPA. Finally, a skeleton line of the stem region was obtained on the basis of the segmentation image, and a picking point positioning algorithm was developed to determine the coordinates of the optimal picking point. Subsequently, the positioning success rate of coordinates, absolute errors, and relative errors were calculated by comparing the automatic positioned coordinates with the manually positioned stem region. Experimental results indicated that this study achieved an average positioning success rate of 92.01%, with an average absolute error of 4.93 pixels and an average relative error of 13.11%. Additionally, the average processing time for processing 640 images using the picking point positioning system is 72.75 ms. This study demonstrates the reliability and effectiveness of positioning mango picking points, laying the technological basis for the automated harvesting of mango fruits.</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"247 ","pages":"Pages 202-220"},"PeriodicalIF":4.4,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142311071","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Laser Doppler vibrometer enables in-situ monitoring of peach firmness","authors":"Dachen Wang ,&nbsp;Yilei Hu ,&nbsp;Jiaqi Xiong ,&nbsp;Yibin Ying ,&nbsp;Ce Yang ,&nbsp;Di Cui","doi":"10.1016/j.biosystemseng.2024.09.013","DOIUrl":"10.1016/j.biosystemseng.2024.09.013","url":null,"abstract":"<div><div>Fruit firmness is a measure of the edible quality and maturity of peaches. In-situ monitoring of peach firmness can aid in fruit quality control and determining the optimal harvest time according to market demand. In this study, a non-contact acoustic vibration-based method was proposed for in-situ monitoring of fruit firmness of on-tree peaches. A new design of a compressed air excitation unit was constructed to impact the peach on the tree and a laser Doppler vibrometer was adopted to measure the acoustic vibration response (AVR) of the peach. To isolate the vibration information characterising fruit firmness, the AVR was firstly pre-processed by the wavelet threshold denoising method and then analysed by the autoregressive method to acquire the power spectral density (PSD) of the peach. For effectively extracting vibration features from the PSD to predict peach firmness, a novel one-dimensional convolutional neural network (CNN_m) with multiscale perceptual fields was constructed. The performance of CNN_m was compared with those of partial least squares regression, support vector regression models, and a single-branch 1D-CNN model with the mean absolute error (MAE), root mean square error (RMSE), coefficient of determination (<em>R</em>^<em>2</em>), and residual prediction deviation (<em>RPD</em>). The results indicated that the proposed method enabled in-situ monitoring of peach firmness and the established CNN_m model performed better than other models in predicting peach firmness (<span><math><mrow><msubsup><mi>R</mi><mi>P</mi><mn>2</mn></msubsup></mrow></math></span> = 0.813, MAEP = 1.636 N mm⁻¹, RMSEP = 2.501 N mm⁻¹, and <span><math><mrow><msub><mrow><mi>R</mi><mi>P</mi><mi>D</mi></mrow><mi>P</mi></msub></mrow></math></span> = 2.334).</div></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":"247 ","pages":"Pages 191-201"},"PeriodicalIF":4.4,"publicationDate":"2024-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142311070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}