Thiago Teixeira Santos

Brazilian Agricultural Research Corporation (EMBRAPA) | Embrapa · Embrapa Digital Agriculture

Functional-structural plant modeling and plant phenotyping require the measurement of geometric features in specimens. This data acquisition is called plant digitizing. Actually, these measurements are performed manually, in invasive or even destructive ways, or using expensive laser scanning equipment. Computer vision based 3D reconstruction is an...

The three-dimensional reconstruction of plants using computer vision methods is a promising alternative to non-destructive metrology in plant phenotyping. However, diversity in plants form and size, different surrounding environments (laboratory, greenhouse or field), and occlusions impose challenging issues. We propose the use of state-of-the-art...

Agricultural applications such as yield prediction, precision agriculture and automated harvesting need systems able to infer the crop state from low-cost sensing devices. Proximal sensing using affordable cameras combined with computer vision has seen a promising alternative, strengthened after the advent of convolutional neural networks (CNNs) as...

Computer vision methods based on convolutional neural networks (CNNs) have presented promising results on image-based fruit detection at ground-level for different crops. However, the integration of the detections found in different images, allowing accurate fruit counting and yield prediction, have received less attention. This work presents a met...

Computer vision methods based on convolutional neural networks (CNNs) have presented promising results on image-based fruit detection at ground-level for different crops. However, the integration of the detections found in different images, allowing accurate fruit counting and yield prediction, have received less attention. This work presents a met...

A short introduction to computer vision applications in agriculture.

The objective of this work was to evaluate different traits of four corn (Zea mays) genotypes with contrasting responses to drought and to determine the main traits associated to such responses. The experiment was carried out in a greenhouse. The plants were grown in pots subjected to full irrigation. Drought was imposed to plants at 54 days after...

The objective of this work was to evaluate different traits of four corn (Zea mays) genotypes with contrasting responses to drought and to determine the main traits associated to such responses. The experiment was carried out in a greenhouse. The plants were grown in pots subjected to full irrigation. Drought was imposed to plants at 54 days after...

Palestra para o evento TechStart Agro Digital 2020

Em uma definição simples e abrangente, visão computacional é o campo da inteligência artificial dedicado à extração de informações a partir de imagens digitais. No contexto da agricultura digital, a visão computacional pode ser empregada na detecção de doenças e pragas, na estimação de safra e na avaliação não invasiva de atributos como qualidade,...

Unmanned aerial vehicles (UAVs) are being increasingly viewed as valuable tools to aid the management of farms. This kind of technology can be particularly useful in the context of extensive cattle farming, as production areas tend to be expansive and animals tend to be more loosely monitored. With the advent of deep learning, and convolutional neu...

Unmanned Aerial Vehicles (UAVs) are being increasingly viewed as valuable tools to aid the management of farms. This kind of technology can be particularly useful in the context of extensive cattle farming, as production areas tend to be expansive and animals tend to be more loosely monitored. With the advent of deep learning, and Convolutional Neu...

A massa de forragem é uma informação de grande importância no manejo de pastagens e de animais em sistemas produtivos. Infelizmente, ainda não existem metodologias bem estabelecidas e fundamentadas para a estimativa dos valores de massa de forragem em larga escala, que permitam sua ampla adoção nos sistemas pecuários e melhorados processos decisóri...

O trabalho descreve uma das experiências do projeto Pecuária do Futuro, que tem como foco a interação com stakeholders e seus três princípios: estratégia integrada de engajamento, influência na governança e geração de inovação. A governança do projeto ocorre em tempo real, com ajustes de direção realizados para chegar ao produto ideal a partir de c...

** Code available at https://github.com/amy-tabb/calico ** Camera network and multi-camera calibration for external parameters is a necessary step for a variety of contexts in computer vision and robotics, ranging from three-dimensional reconstruction to human activity tracking. This paper describes CALICO, a method for camera network and/or multi-...

Introdução ao uso de visão computacional para extração de traits em fenotipagem por imagem.

RESUMO: O registro da arquitetura de uma planta, isto é, a descrição da topologia e da geometria de sua estrutura, é empregado em estudos funcionais-estruturais variados como intercepção solar, interações bióticas e abióticas e dispersão de defensivos, entre outros. Porém, a realização desse registro é extremamente laboriosa, especialmente em plant...

Agricultural applications as yield prediction, precision agriculture and automated harvesting need systems able to infer the culture state from low cost sensing devices. Proximal sensing using affordable cameras combined to computer vision have seen a promising alternative, strengthened after the advent of convolutional neural networks (CNNs) as an...

Automação em ambientes internos e urbanos geralmente depende de ambientes estruturados ou semiestruturados. Das linhas de montagem industriais robotizadas à navegação automotiva ciente do tráfego urbano, algum conhecimento da estrutura do ambiente é empregado na prestação de serviços automatizados. No entanto, os ambientes naturais e agrícolas care...

The feasibility of thermography as a technique for plant screening aiming at drought-tolerance has been proven by its relationship with gas exchange, biomass, and yield. In this study, unlike most of the previous, thermography was applied for phenotyping contrasting maize genotypes whose classification for drought tolerance had already been establi...

In this work, we modeled the problem of detection of fruit and leaves in viticulture for proximal applications as a supervised machine learning task. We created and manually labeled a database of images obtained at Guaspari Winery. In total, the database consists of 11.883 images of bunch of grapes and leaves. We trained a convolutional network wit...

Uma introdução ao uso de IoT no campo, com exemplos de pilotos e aplicações.

In this work, we modeled the problem of detection of fruit and leaves in viticulture for proximal applications as a supervised machine learning task. We created and manually labeled a database of images obtained in April 2017 at Guaspari Winery. In total, the database consists of 11,883 images of bunch of grapes and leaves. We trained a convolution...

Over the last few years, considerable effort has been spent by Embrapa in the construction of a plant disease database representative enough for the development of effective methods for automatic plant disease detection and recognition. In October of 2016, this database, called PDDB, had 2326 images of 171 diseases and other disorders affecting 21...

Breve discussão sobre métodos quantitativos e computação científica (MQCC) aplicados a dados obtidos por automação em agricultura, com ênfase em fotogrametria e percepção em visão computacional.

O presente trabalho apresenta uma metodologia para a análise fenotı́pica de vinhas por re- construção 3-D, a partir de imagens obtidas por uma webcam de alta definição e baixo custo. Um novo software aplicativo integrou componentes de odometria visual e visão estéreo por múltiplas imagens para criar nuvens de pontos tridimensionais densas e precisa...

This work presents a methodology for 3-D phenotyping of vineyards based on images captured by a low cost high-definition webcamera. A novel software application integrated visual odometry and multiple-view stereo components to create dense and accurate three-dimensional points clouds for vines, properly transformed to millimeter scale. Geometrical...

Cenário do uso de IoT no campo e perspectivas.

Esta apresentação aborda a IoT na agricultura, suas aplicações, desafios e alguns resultados já reportados pela academia e pelo mercado.

The gap between the current capabilities of image-based methods for automatic plant disease identification and the real-world needs is still wide. Although advances have been made on the subject, most methods are still not robust enough to deal with a wide variety of diseases and plant species. This paper proposes a method for disease identificatio...

Visão computacional consiste em inferir características do mundo físico a partir de imagens digitais. Câmeras digitais são sensores baratos e bem disseminados, de modo que sistemas de visão computacional podem ser alternativas de baixo custo para problemas de sensoriamento, como os encontrados na agricultura. Porém, a agricultura apresenta contexto...

A method for automatic detection of coffee berries in digital images acquired in the field is presented. It consists of a preliminary detection step that finds candidate regions, followed by a classification stage which filters out false positives. For training and evaluation of automatic classification, 3,393 images were manually labeled. Quantita...

In research and development (R&D), interactive computing environments are a frequently employed alternative for data exploration, algorithm development and prototyping. In the last twelve years, a popular scientific computing environment flourished around the Python programming language. Most of this environment is part of (or built over) a softwar...

Functional-structural modeling and high-throughput phenomics demand tools for 3D measurements of plants. In this work, structure from motion is employed to estimate the position of a hand-held camera, moving around plants, and to recover a sparse 3D point cloud sampling the plants’ surfaces. Multiple-view stereo is employed to extend the sparse mod...

Functional-structural modeling and high-throughput pheno-mics demand tools for 3D measurements of plants. In this work, struc-ture from motion is employed to estimate the position of a hand-held camera, moving around plants, and to recover a sparse 3D point cloud sampling the plants' surfaces. Multiple-view stereo is employed to extend the sparse m...

A stereo approach for 3D plant modelling is presented. Using only a set of photographies, the method produces a dense 3D point cloud that samples the plant surface. Clustering automatically segments the plant structure into meaningful parts, which are classified as leaves or internodes. Measurements can be computed for each element, as area or surf...

Functional-structural plant modeling and plant phe- notyping require the measurement of geometric features in spec- imens. This data acquisition is called plant digitizing. Actually, these measurements are performed manually, in invasive or even destructive ways, or using expensive laser scanning equipment. Computer vision based 3D reconstruction i...

This paper introduces the Laboratory of Technologies for Interaction(LaTIn) and briefly describes its current main projects. The mainfocus of LaTInhas been developing new ways of human-machineinteraction using computer vision techniques. The projects arecathegorized according to the distance between the human user and themachine being operated. For...

Automated virtual camera control has been widely used in animation and interactive virtual environments. We have developed a multiple sparse camera based free view video system prototype that allows users to control the position and orientation of a virtual camera, enabling the observation of a real scene in three dimensions (3D) from any desired v...

This paper proposes a method to locate and track people by combining evidence from multiple cameras using the homography constraint. The proposed method use foreground pixels from simple background subtraction to compute evidence of the location of people on a reference ground plane. The algorithm computes the amount of support that basically corre...

Free Viewpoint Video (FVV) might become one of the most interesting features in future digital television, enabling viewers to move around freely and observe the scene in three dimensions (3D) from any desired viewpoint, similar to many 3D interactive computer graphics applications today. This paper describes a FVV appropriate for virtual reality s...

This paper proposes a method to locate people on a reference plane using multiple cameras. Previous works rely on people trajectories and color models to solve occlusion.This new approach solves people detection under occlusion by accumulating evidence from multiple views instantaneously and does not rely on previous segmentation of individuals in...

This paper introduces a fast, robust and accurate iris segmentation technique based on active lighting. The geometry of the light sources, a single camera and the eye facilitates the detection of the pupil and the automatic selection of the most appropriate image for biometric identification from the video stream, minimizing the effects of noise, d...

This paper presents a framework to solve the video shot segmentation problem. Temporal video segmentation is the first step in the process of indexing, browsing and searching digital video information. The segmentation is based on a decision function which defines if two frames belong to the same shot or not. Three methods are discussed: pair-wise...

Video contains huge amounts of information which needs to be structured and indexed to make tasks like search and navigation possible. This paper describes ongoing research for the development of a prototype system to segment digital video in shots and index them by key-frames. Preliminary results of videos structured as a sequence of key-frames is...

Resumo: Os constantes avanços em poder de processamento, capacidade de armazenamento, e largura de banda tornaram possível ao usuário de com- putadores o acesso, manipulação e edição de imagens e vídeos digitais. Nesse tutorial, vamos descrever alguns algoritmos de visão computacional para a seg- mentação, indexação e recuperação de vídeo utilizand...

This paper provides a compact overview of LaTIn shot boundary detection task system and results (LaTIn-USP MD rev107) for the TRECVID 2006. The run is an image processing approach that has 2 steps: an event detec-tor that performs absolute pixel difference between each frame and a median frame followed by a shot boundary de-tection step based on hi...

This paper describes an academic research project in com-puter vision and distance learning currently under development in Brazil, and it can be roughly classified in the 1st workshop area, about com-puter vision research that addresses educational problems in developing regions. In particular, we describe our experience in developing two com-puter...