RESUMO
Introduction: Precise semantic segmentation of microbial alterations is paramount for their evaluation and treatment. This study focuses on harnessing the SegFormer segmentation model for precise semantic segmentation of strawberry diseases, aiming to improve disease detection accuracy under natural acquisition conditions. Methods: Three distinct Mix Transformer encoders - MiT-B0, MiT-B3, and MiT-B5 - were thoroughly analyzed to enhance disease detection, targeting diseases such as Angular leaf spot, Anthracnose rot, Blossom blight, Gray mold, Leaf spot, Powdery mildew on fruit, and Powdery mildew on leaves. The dataset consisted of 2,450 raw images, expanded to 4,574 augmented images. The Segment Anything Model integrated into the Roboflow annotation tool facilitated efficient annotation and dataset preparation. Results: The results reveal that MiT-B0 demonstrates balanced but slightly overfitting behavior, MiT-B3 adapts rapidly with consistent training and validation performance, and MiT-B5 offers efficient learning with occasional fluctuations, providing robust performance. MiT-B3 and MiT-B5 consistently outperformed MiT-B0 across disease types, with MiT-B5 achieving the most precise segmentation in general. Discussion: The findings provide key insights for researchers to select the most suitable encoder for disease detection applications, propelling the field forward for further investigation. The success in strawberry disease analysis suggests potential for extending this approach to other crops and diseases, paving the way for future research and interdisciplinary collaboration.
RESUMO
Egypt is among the world's largest producers of sugarcane. This crop is of great economic importance in the country, as it serves as a primary source of sugar, a vital strategic material. The pre-cutting planting mode is the most used technique for cultivating sugarcane in Egypt. However, this method is plagued by several issues that adversely affect the quality of the crop. A proposed solution to these problems is the implementation of a sugarcane-seed-cutting device, which incorporates automatic identification technology for optimal efficiency. The aim is to enhance the cutting quality and efficiency of the pre-cutting planting mode of sugarcane. The developed machine consists of a feeding system, a node scanning and detection system, a node cutting system, a sugarcane seed counting and monitoring system, and a control system. The current research aims to study the pulse widths (PW) of three-color channels (R, G, and B) of the RGB color sensors under laboratory conditions. The output PW of red, green, and blue channel values were recorded at three color types for hand-colored nodes [black, red, and blue], three speeds of the feeding system [7.5 m/min, 5 m/min, and 4.3 m/min], three installing heights of the RGB color sensors [2.0 cm, 3.0 cm, and 4.0 cm], and three widths of the colored line [10.0 mm, 7.0 mm, and 3.0 mm]. The laboratory test results s to identify hand-colored sugarcane nodes showed that the recognition rate ranged from 95% to 100% and the average scanning time ranged from 1.0 s to 1.75 s. The capacity of the developed machine ranged up to 1200 seeds per hour. The highest performance of the developed machine was 100% when using hand-colored sugarcane stalks with a 10 mm blue color line and installing the RGB color sensor at 2.0 cm in height, as well as increasing the speed of the feeding system to 7.5 m/min. The use of IoT and RGB color sensors has made it possible to get analytical indicators like those achieved with other automatic systems for cutting sugar cane seeds without requiring the use of computers or expensive, fast industrial cameras for image processing.
