Noé Velázquez-López
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Noé Velázquez-López has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Gilberto de Jesús López-Canteñs
Posgrado en Ingeniería Agrícola y Uso Integral del Agua, Universidad Autónoma Chapingo, Carretera México-Texcoco km 38.5, Chapingo 56230, Mexico
2 shared publications - Alan Hernández-Mercado1 shared publication
- Alvaro Murguia-Cozar
Posgrado en Ingeniería Agrícola y Uso Integral del Agua, Universidad Autónoma Chapingo, Carretera México-Texcoco km 38.5, Chapingo 56230, Mexico
1 shared publication - Antonia Macedo-Cruz
Postgrado de Hidrociencias, Colegio de Postgraduados, Carretera México-Texcoco km. 36.5, Montecillo 56264, Mexico
1 shared publication - Carlos Martinez-Nicolas1 shared publication
- David I. Sánchez-Chávez
Graduate Program in Agricultural Engineering and Integral Use of Water, Chapingo Autonomous University, Carretera Federal México-Texcoco km. 38.5, Texcoco C.P. 56230, Estado de México, Mexico
1 shared publication
Latest Publications
- Open Access
CNN-Based Temporal Fusion of UAV and Sentinel-2 Imagery for Crop Monitoring
The use of Sentinel-2 satellite imagery for crop monitoring is constrained by its native spatial resolution. To address this limitation, data fusion-based super-resolution techniques have emerged as a promising approach for generating imagery with enhanced spatial detail. In this study, we propose TVDSR, a Sentinel-2 and UAV image fusion model that integrates temporal and spectral information through a Bidirectional ConvLSTM-based architecture. The images used for super-resolution were constructed from spectral compositions of five vegetation indices. Model performance was evaluated using 15 acquisition dates, including 11 Sentinel-2 images acquired in 2024 and 4 acquired in 2025. Two independent dates were reserved for validation using reconstruction quality metrics. The proposed model was compared against three state-of-the-art reference models: S3-ESRGAN, DeepSent, and UAV-Net. UAV-Net achieved the best performance for the 2024 validation image (RMSE = 0.128, MAE = 0.10, PSNR = 24.06 dB, and SSIM = 0.86). However, TVDSR obtained the best results for the 2025 evaluation image (RMSE = 0.149, MAE = 0.115, PSNR = 22.60 dB, and SSIM = 0.85) and exhibited the lowest performance degradation across acquisition dates, indicating superior temporal generalization capability. Furthermore, TVDSR more consistently preserved the derived vegetation indices, reflecting a better retention of spectral information associated with crop condition, while producing reconstructions that were visually closer to the reference UAV imagery. The joint integration of temporal and spectral information constitutes an effective strategy for improving the robustness and generalization capability of Sentinel-2 image super-resolution for crop monitoring, precision agriculture, crop condition assessment, and agricultural management applications.
Open AccessCNN-Based Temporal Fusion of UAV and Sentinel-2 Imagery for Crop Monitoring
The use of Sentinel-2 satellite imagery for crop monitoring is constrained by its native spatial resolution. To address this limitation, data fusion-based super-resolution techniques have emerged as a promising approach for generating imagery with enhanced spatial detail. In this study, we propose TVDSR, a Sentinel-2 and UAV image fusion model that integrates temporal and spectral information through a Bidirectional ConvLSTM-based architecture. The images used for super-resolution were constructed from spectral compositions of five vegetation indices. Model performance was evaluated using 15 acquisition dates, including 11 Sentinel-2 images acquired in 2024 and 4 acquired in 2025. Two independent dates were reserved for validation using reconstruction quality metrics. The proposed model was compared against three state-of-the-art reference models: S3-ESRGAN, DeepSent, and UAV-Net. UAV-Net achieved the best performance for the 2024 validation image (RMSE = 0.128, MAE = 0.10, PSNR = 24.06 dB, and SSIM = 0.86). However, TVDSR obtained the best results for the 2025 evaluation image (RMSE = 0.149, MAE = 0.115, PSNR = 22.60 dB, and SSIM = 0.85) and exhibited the lowest performance degradation across acquisition dates, indicating superior temporal generalization capability. Furthermore, TVDSR more consistently preserved the derived vegetation indices, reflecting a better retention of spectral information associated with crop condition, while producing reconstructions that were visually closer to the reference UAV imagery. The joint integration of temporal and spectral information constitutes an effective strategy for improving the robustness and generalization capability of Sentinel-2 image super-resolution for crop monitoring, precision agriculture, crop condition assessment, and agricultural management applications.
- Open Access
- 25 August 2026
- Published by Editorial CSIC in Spanish Journal of Agricultural Research
- Vol. 24 (2)
Detection of flowering in pineapple (Ananas comosus) using deep learning and drone technology
Aim of study: Identifying flowering in pineapples is critical for the selective application of growth regulators and for achieving a uniform harvest. The objective of this study was to develop an efficient and accurate method for detecting flowering in pineapples (Ananas comosus) using drones and state-of-the-art deep learning models. Area of study: The research was conducted on 1.8 ha of agricultural plots provided by the company DIMASUR, located in Juan Rodríguez Clara, Veracruz, Mexico. Material and methods: A Mavic 3 drone, equipped with a Hasselblad camera, was used to capture images at an altitude of 14 meters, obtaining a nominal GSD of 0.38 cm/pixel. A robust dataset of 6,116 standardized images (640 x 640 tiles) and approximately 143,500 labels was generated under various lighting conditions, after applying tiling and data augmentation techniques. The performance of YOLOv8n, YOLOv10n, and YOLOv11n architectures was evaluated and compared using a unified set of hyperparameters (Adam optimizer, lr0 = 0.01) to ensure a fair comparison. Main results: The YOLOv11n model achieved the best overall performance, with an accuracy of 95.95%, a recall of 93.69%, a precision of 97.47%, and an mAP@0.5-0.95 of 80.47%. This model demonstrated superior stability in the face of occlusions and flower overlap, while also achieving the most efficient training time (5.6 hours). These results confirm that YOLOv11n offers the optimal balance between accuracy and sensitivity, while maintaining low computational costs ideal for real-time applications. Research highlights: The integration of high-resolution drones and YOLO algorithms enables flowering detection with an accuracy of over 97%. (1) The YOLOv11n model proved to be more reliable and consistent than its predecessors in the face of variations in lighting and crop density. (2) The proposed approach optimizes agricultural monitoring, enabling data-driven decision-making that maximizes profitability and operational efficiency in pineapple cultivation.
- 25 August 2026
- Published by Editorial CSIC in Spanish Journal of Agricultural Research
- Vol. 24 (2)
Open AccessDetection of flowering in pineapple (Ananas comosus) using deep learning and drone technology
Aim of study: Identifying flowering in pineapples is critical for the selective application of growth regulators and for achieving a uniform harvest. The objective of this study was to develop an efficient and accurate method for detecting flowering in pineapples (Ananas comosus) using drones and state-of-the-art deep learning models. Area of study: The research was conducted on 1.8 ha of agricultural plots provided by the company DIMASUR, located in Juan Rodríguez Clara, Veracruz, Mexico. Material and methods: A Mavic 3 drone, equipped with a Hasselblad camera, was used to capture images at an altitude of 14 meters, obtaining a nominal GSD of 0.38 cm/pixel. A robust dataset of 6,116 standardized images (640 x 640 tiles) and approximately 143,500 labels was generated under various lighting conditions, after applying tiling and data augmentation techniques. The performance of YOLOv8n, YOLOv10n, and YOLOv11n architectures was evaluated and compared using a unified set of hyperparameters (Adam optimizer, lr0 = 0.01) to ensure a fair comparison. Main results: The YOLOv11n model achieved the best overall performance, with an accuracy of 95.95%, a recall of 93.69%, a precision of 97.47%, and an mAP@0.5-0.95 of 80.47%. This model demonstrated superior stability in the face of occlusions and flower overlap, while also achieving the most efficient training time (5.6 hours). These results confirm that YOLOv11n offers the optimal balance between accuracy and sensitivity, while maintaining low computational costs ideal for real-time applications. Research highlights: The integration of high-resolution drones and YOLO algorithms enables flowering detection with an accuracy of over 97%. (1) The YOLOv11n model proved to be more reliable and consistent than its predecessors in the face of variations in lighting and crop density. (2) The proposed approach optimizes agricultural monitoring, enabling data-driven decision-making that maximizes profitability and operational efficiency in pineapple cultivation.
- Open Access
- 6 May 2025
- Published by MDPI AG in Agriengineering
- Vol. 7 (5) , 138
Physical–Mechanical Properties of Tomato Seedlings for the Design and Optimization of Automatic Transplanters
This study was based on the hypothesis that the hybrid type and its physical–mechanical properties significantly influence the operational efficiency of transplanting systems. Understanding these properties is essential for optimizing the performance of semi-automatic and automatic transplanters. To test this hypothesis, a completely randomized design was implemented to evaluate the physical–mechanical properties of tomato seedlings. A total of 1350 seedlings from three F1 hybrids—Natalie (H1), CID (H2), and Gavilán (H3)—cultivated in central Mexico, were analyzed. The statistical analyses included mean comparisons using Tukey’s test and multiple linear regression to estimate the center of mass (CM). The results indicate that H2 was notable for its total height ( = 311.76 mm), canopy development in X, Y, and Z axes (170.24 mm, 106.84 mm, and 98.14 mm, respectively), stem diameter ( = 3.65 mm), total weight ( = 11.92 g), (78.36 mm) and (233.40 mm) distances, and oscillation period (T = 0.88 s). H1 had the highest stem height ( = 53.18 mm), = 11.76 g, and root ball (RB) moisture content (MC) (77.36%). H3 had the largest = 3.70 mm, as well as the highest MC in the stem (94.51%) and the remaining foliage (92.92%). Regarding mechanical properties, the average adhesion force (AF) was 4.606 N (H1), 7.470 N (H2), and 3.815 N (H3). The average root ball punching force (RBPF) was 0.36, 0.48, and 0.25 N, respectively. The lowest static friction coefficient (SFC) on a galvanized steel sheet was 0.936. The drop test (DT) revealed an average residual substrate mass of 0.148 g at a height of 500 mm. It can be concluded that the interaction between hybrid type, transplanting age, and MC plays a critical role in the efficient design of semi-automatic and automatic transplanting equipment. This interaction enables process optimization, ensures operational quality, reduces seedling damage, and ultimately enhances and increases the long-term profitability and sustainability of the equipment.- 6 May 2025
- Published by MDPI AG in Agriengineering
- Vol. 7 (5) , 138
Open AccessPhysical–Mechanical Properties of Tomato Seedlings for the Design and Optimization of Automatic Transplanters
This study was based on the hypothesis that the hybrid type and its physical–mechanical properties significantly influence the operational efficiency of transplanting systems. Understanding these properties is essential for optimizing the performance of semi-automatic and automatic transplanters. To test this hypothesis, a completely randomized design was implemented to evaluate the physical–mechanical properties of tomato seedlings. A total of 1350 seedlings from three F1 hybrids—Natalie (H1), CID (H2), and Gavilán (H3)—cultivated in central Mexico, were analyzed. The statistical analyses included mean comparisons using Tukey’s test and multiple linear regression to estimate the center of mass (CM). The results indicate that H2 was notable for its total height ( = 311.76 mm), canopy development in X, Y, and Z axes (170.24 mm, 106.84 mm, and 98.14 mm, respectively), stem diameter ( = 3.65 mm), total weight ( = 11.92 g), (78.36 mm) and (233.40 mm) distances, and oscillation period (T = 0.88 s). H1 had the highest stem height ( = 53.18 mm), = 11.76 g, and root ball (RB) moisture content (MC) (77.36%). H3 had the largest = 3.70 mm, as well as the highest MC in the stem (94.51%) and the remaining foliage (92.92%). Regarding mechanical properties, the average adhesion force (AF) was 4.606 N (H1), 7.470 N (H2), and 3.815 N (H3). The average root ball punching force (RBPF) was 0.36, 0.48, and 0.25 N, respectively. The lowest static friction coefficient (SFC) on a galvanized steel sheet was 0.936. The drop test (DT) revealed an average residual substrate mass of 0.148 g at a height of 500 mm. It can be concluded that the interaction between hybrid type, transplanting age, and MC plays a critical role in the efficient design of semi-automatic and automatic transplanting equipment. This interaction enables process optimization, ensures operational quality, reduces seedling damage, and ultimately enhances and increases the long-term profitability and sustainability of the equipment. - Open Access
Design and Comparison of Two Maize Seeders Coupled with an Agricultural Robot
In recent years, the development of robotic vehicles in agriculture has made it possible to reduce human intervention and fatigue in carrying out arduous or repetitive tasks, as well as helping to promote sustainable agriculture to address climate change. However, the great diversity of agricultural tasks and the varied production systems and crops demand a wide range of solutions that can be adapted to robotic vehicles as a power source. These alternatives must be affordable and user-friendly for some users, although more sophisticated solutions must also be developed for others, depending on their specific needs. For this, the present work focuses on the development of two maize seeders with different metering systems coupled to an agricultural robot. The first seeder has a conventional mechanically driven seed metering system with a drive wheel and chain gear, while the second one has an electronically driven metering system based on a DC motor and a digital encoder controlled by a microcontroller. Both seeders were coupled to a remote-controlled robotic vehicle and evaluated on real farmland. Seed distribution in the seed rows was contrasting; the results indicated that the mechanical system performed better in the field than the electronic system. For both seeders, the working capacity was approximately 0.135 ha/h at an average speed of 2.0 km/h. The proposed robot–seeder assembly could help farmers automate and reduce the workload associated with planting, as well as attract young people to the field.
Open AccessDesign and Comparison of Two Maize Seeders Coupled with an Agricultural Robot
In recent years, the development of robotic vehicles in agriculture has made it possible to reduce human intervention and fatigue in carrying out arduous or repetitive tasks, as well as helping to promote sustainable agriculture to address climate change. However, the great diversity of agricultural tasks and the varied production systems and crops demand a wide range of solutions that can be adapted to robotic vehicles as a power source. These alternatives must be affordable and user-friendly for some users, although more sophisticated solutions must also be developed for others, depending on their specific needs. For this, the present work focuses on the development of two maize seeders with different metering systems coupled to an agricultural robot. The first seeder has a conventional mechanically driven seed metering system with a drive wheel and chain gear, while the second one has an electronically driven metering system based on a DC motor and a digital encoder controlled by a microcontroller. Both seeders were coupled to a remote-controlled robotic vehicle and evaluated on real farmland. Seed distribution in the seed rows was contrasting; the results indicated that the mechanical system performed better in the field than the electronic system. For both seeders, the working capacity was approximately 0.135 ha/h at an average speed of 2.0 km/h. The proposed robot–seeder assembly could help farmers automate and reduce the workload associated with planting, as well as attract young people to the field.
- Open Access
- 9 July 2024
- Published by University of Maribor in Agricultura Scientia
- Vol. 21 (1) , 35-46
Robot for Navigation in Maize Crops for the Field Robot Event 2023
Navigation in a maize crop is a crucial task for the development of autonomous robots in agriculture, with numerous applications such as spraying, monitoring plant growth and health, and detecting weeds and pests. The Field Robot Event 2023 (FRE) continued to challenge universities and other research teams to push the development of algorithms for agricultural robots further. The Universidad Autónoma Chapingo has been developing a robot for various agricultural tasks, aiming to provide a low-cost alternative to work with Mexican farmers in the future. For this edition of the FRE, a navigation algorithm was created using an encoder, an IMU (Inertial Measurement Unit), an RPLIDAR (Rotating Platform Light Detection and Ranging), and cameras to collect data for decision-making. The algorithm was developed in ROS Melodic, dividing the task into steps that were tested to determine the robot's actual movements. The system navigates by using ROIs (regions of interest) and the mass center to guide the robot between maize rows. It calculates the mean of the final orientation values before reaching the end of a row, which is detected using an RPLIDAR. For turns and straight-line movements to reach the next row, the orientation is used as a guide. To detect plants for spraying, lasers located on each side of the vehicle are employed. Obstacle detection relies on a YOLOv5 (You Only Look Once) trained model and a laser, while reverse navigation uses a rear camera. During the competition, the robot faced challenges such as dealing with grass, the small size of the plants, and the need to use a different power source, which affected its performance.
- 9 July 2024
- Published by University of Maribor in Agricultura Scientia
- Vol. 21 (1) , 35-46
Open AccessRobot for Navigation in Maize Crops for the Field Robot Event 2023
Navigation in a maize crop is a crucial task for the development of autonomous robots in agriculture, with numerous applications such as spraying, monitoring plant growth and health, and detecting weeds and pests. The Field Robot Event 2023 (FRE) continued to challenge universities and other research teams to push the development of algorithms for agricultural robots further. The Universidad Autónoma Chapingo has been developing a robot for various agricultural tasks, aiming to provide a low-cost alternative to work with Mexican farmers in the future. For this edition of the FRE, a navigation algorithm was created using an encoder, an IMU (Inertial Measurement Unit), an RPLIDAR (Rotating Platform Light Detection and Ranging), and cameras to collect data for decision-making. The algorithm was developed in ROS Melodic, dividing the task into steps that were tested to determine the robot's actual movements. The system navigates by using ROIs (regions of interest) and the mass center to guide the robot between maize rows. It calculates the mean of the final orientation values before reaching the end of a row, which is detected using an RPLIDAR. For turns and straight-line movements to reach the next row, the orientation is used as a guide. To detect plants for spraying, lasers located on each side of the vehicle are employed. Obstacle detection relies on a YOLOv5 (You Only Look Once) trained model and a laser, while reverse navigation uses a rear camera. During the competition, the robot faced challenges such as dealing with grass, the small size of the plants, and the need to use a different power source, which affected its performance.