Image Processing in Floriculture Using a robotic Mobile Platform

Image Processing in Floriculture Using a robotic Mobile Platform

García Torres Juan Mauricio Fundación Universitaria Agraria De Colombia
Bogotá, Colombia juang1145@hotmail.com

Caro Prieto Diana Carolina Fundación Universitaria Agraria De Colombia
Bogotá, Colombia carodiana3@gmail.com

Abstract— Colombia has a privileged geographical location which makes it a cornerstone and equidistant point to all regional markets. The country has a great ecological diversity and it’s one of the largest suppliers of flowers for US. Colombian flower companies have made innovations in the marketing process, using methods to reach all conditions for final consumers. Nowadays, Robotic has performed a significant role in agriculture thanks to its the level of accuracy. Examples of this are all developments in computer vision, allowing greater control over production processes such as quality inspection, selection and classification of a product according to market requirements, ensuring high quality standards in production for exporting, optimizing performance times at the production level, among others.
Computer vision is based on image processing, method used for identifying environment objects and making decisions about their behavior. The main tasks of the vision system are detecting, segmenting and recognizing patterns, integrating systems of digital image capture, input-output devices and computer networks.
The image processing in real time using artificial vision systems applied in floriculture is mainly focused on the use of techniques and methods to obtain, process and analyze the information provided by the images of different types of flowers to export. In this case, fulfilling the goal of quality inspection system checks the compliance of a flower with certain characteristics in their appearance, such as size, uniformity of color, brightness, etc.
This article develops a monitoring system for floriculture industries. The system was implemented in a robotic platform. This device has the ability to be programmed in different programming languages. The robot takes the necessary environment information from its camera. The algorithm of the monitoring system was developed with the image processing toolbox on Matlab. The implemented algorithm acquires images through its camera, it performs a pre- processing of the image, noise filter, enhancing of the color and adjusting the dimension in order to increase processing speed. Then, the image is segmented by color and with the binarized version of the image using morphological operations (erosion and dilation), extract relevant features such as centroid, perimeter and area. The data obtained from the image processing helps the robot with the automatic identification of objectives, orientation and move towards them. Also, the results generate a diagnostic quality of each object scanned.

Keywords—image proccesing, morphologic operations, floriculture. I. INTRODUCTION
The flower industry is undoubtedly one of the most successful businesses in the country; it’s the third largest foreign exchange earner after oil and coffee. It has been an important employer generator in Colombia, evidenced over 30 years. The Colombian floriculture has exceeded sales up more than $ 500 million annual [1], which has led Colombia today be the second largest exporter of flowers in the world after The Netherlands [2].

However, this industry has several problems: the lack of oversight on the product selection since the harvest period depending on the features required to export, the use of high residual substances which causes pollution in the setting and early discoloration petals [3]. Other issues mentioned are the control of temperature, humidity and pests, but these depend on the environment where they were developed.

A possible solution to increase the quality of the product exported to other countries consists on applying the technology on the agriculture sector. They have done different studies regarding the application of image processing in agriculture on coffee [4].

Also, another study was done using morphological operations in the coffee with three different methods using the technical of region’s growth. The result suggested on the reseach recommends the implementation of systems with discontinuity information, pixel by pixel and region-and contour [5]. 2

Some process have been made using PLC to classify different objects depending on the color and material, using optical and inductive sensors [6].

Several industrial applications made on products like coffee, have also been studied in flowers, focused on their behaviors such as the way they open and close their petals, growth and their adaptation in the environment [7].

In base on the background studied, we propose the use of automation in floriculture, because companies can ensure the quality on the produc

…(Full text truncated)…