Iberoamerican Journal of Applied Computing

USE OF COMPUTER VISION FOR INSECT RECOGNITION

Fri, 22 Nov 2024 00:00:00 +0000

The use of technologies such as computer vision has been a trend in agriculture. This technology can reduce the effort and time to obtain data about crops, improving production, crop health and bringing more sustainable practices. This article presents a literature review, among the topics covered, the article includes algorithms and computational techniques, including Convolutional Neural Networks (CNNs), YOLO (You Only Look Once) and Support Vector Machines (SVMs). These techniques are used in conjunction with various image processing techniques to detect and classify insects. Machine learning techniques have been shown to be effective in classifying insects based on characteristics such as color, size and position. Different datasets, such as IP102, Pest24, Xie1, Xie2 and Wang, were used to train and evaluate these algorithms. This article concludes the effectiveness of deep learning algorithms, particularly YOLOv5 and Faster R-CNN, in insect detection and classification, suggesting a promising future for automated insect monitoring.

ASPHALT PAVEMENT CRACKING CLASSIFICATION USING CONVOLUTIONAL NEURAL NETWORKS

Luciano Jose Senger, Bruno Seixas Bonfati, Lilian Tais de Gouveia — Tue, 16 Jul 2024 00:00:00 +0000

Asphalt pavements on roads are subject to cracking due mainly to overload and weather over time. Detecting the presence of cracking is an essential part of road maintenance systems. Traditional road defect detection methods are time-consuming, dangerous, labor-intensive, and subjective. An alternative is to use digital images of roads, collected by cars or unmanned aerial vehicles, to feed automated asphalt damage identification models. Thus, automated defect detection systems could quantify the quality of road surfaces and help prioritize and plan road network maintenance, thereby preserving and extending the useful life of roads. This work aims to investigate the use of machine learning for detecting and classifying cracking in road pavements using convolutional neural networks. Transfer learning and data augmentation techniques were used to build classification models. The results indicated that the model was able to identify cracking in the images of roads with an accuracy of 99%. When using the model to classify the cracking subtypes, the model presented an overall accuracy of 78%.

USE OF UAV AS A TOOL FOR DAM AUSCULTATION

Tue, 03 Dec 2024 00:00:00 +0000

This study explores the application of Unmanned Aerial Vehicles (UAVs) integrated with digital aerophotogrammetry as a tool for the auscultation of earth dams, focusing on the Chasqueiro Dam, located in southern Rio Grande do Sul. Using a rotary-wing UAV equipped with a high-resolution camera, cartographic products such as orthomosaics and Digital Surface Models (DSM) were generated, enabling the identification of critical elements such as piezometers and gullies on the downstream slope that were not detected during visual inspections. The data analysis allowed for the evaluation of the variability of slope inclinations and the conditions of the dam crest, confirming that its width and elevation are adequate for structural stability. The use of techniques such as georeferencing through materialized control points and processing with specialized software demonstrated high precision and efficiency in monitoring. This method proved to be a safe, fast, and cost-effective alternative compared to conventional approaches. It is concluded that employing advanced technologies in dams can mitigate the risks of structural failures and significantly improve monitoring and maintenance practices. Continued research is recommended to integrate additional sensors to expand and refine structural and environmental analyses.

INTEGRATION OF RFID TECHNOLOGY IN AGRICULTURAL TRACEABILITY TO OPTIMIZE CONTROL AND QUALITY OF PRODUCTS

Mon, 12 Aug 2024 00:00:00 +0000

Traceability in the agricultural chain is essential to guarantee the safety and quality of food products. The use of technologies such as Radio Frequency Identification (RFID) presents itself as a solution to improve this process, providing accurate and efficient tracking of products throughout the entire production chain. This study aimed to develop and validate an integrated structure using RFID in the traceability process in the soybean grain production chain. The results of this work included the detailed specification of the stages of the production process, using RFID for tracking and monitoring. The integration of RFID proved to be effective, allowing more accurate and efficient traceability, ensuring compliance with quality standards and industry regulations.

PLANT SPECIES RECOGNITION USING LEAF IMAGES AND CONVOLUTIONAL NEURAL NETWORKS

Wed, 29 May 2024 00:00:00 +0000

The identification of plant species is essential in botany and has attracted the interest of researchers in the field of computer science. Such identification requires the assistance of botany experts and, due to the substantial number of species and the similarity among them, it can be time-consuming and subjective. To automate the process of plant identification, computer systems that capture and process plant images have been considered. These systems use machine learning and therefore require image samples for training and model construction. Among the techniques that can be used for machine learning, convolutional neural networks have shown promise due to their ability to use images without prior preprocessing and background information. This work investigates the use of machine learning through convolutional neural networks to identifying plant species. For this, a new dataset of images from 35 plant species were created, collecting images from an arboreal collection, and, using data augmentation, this dataset was expanded. This dataset was used to evaluate the accuracies of four convolutional neural network models. The better accuracy value was equal to 89%, when using the MobileNetV2 model.