Sidnei Alves de Araújo

Ciratefi: An RST-invariant template matching with extension to color images

Template matching is a technique widely used for finding patterns in digital images. A good template matching should be able to detect template instances that have undergone geometric transformations. In this paper, we proposed a grayscale template matching algorithm named Ciratefi, invariant to rotation, scale, translation, brightness and contrast and its extension to color images. We introduce CSSIM (color structural similarity) for comparing the similarity of two color image patches and use it in our algorithm. We also describe a scheme to determine automatically the appropriate parameters of our algorithm and use pyramidal structure to improve the scale invariance. We conducted several experiments to compare grayscale and color Ciratefis with SIFT, C-color-SIFT and EasyMatch algorithms in many different situations. The results attest that grayscale and color Ciratefis are more accurate than the compared algorithms and that color-Ciratefi outperforms grayscale Ciratefi most of the time. However, Ciratefi is slower than the other algorithms.

Optimization of vacuum cooling treatment of postharvest broccoli using response surface methodology combined with genetic algorithm technique

Abstract In this paper, the effects of vacuum cooling factors on the weight loss of postharvest broccoli were initially investigated. In sequence, the vacuum cooling treatment conditions were optimized using the response surface methodology (RSM) combined with the genetic algorithm (GA) technique. Fresh broccoli samples were harvested from a Chinese farm, and the green heads of selected samples were cut into smaller pieces, with diameters approximately 3–4 cm, and sequentially equilibrated to room temperature. Pressure (200–600 Pa), broccoli weight (200–500 g), water volume (2–6%, v/v) and time (20–40 min) were used as factors, and weight loss and end temperature were recorded as responses. The GA was employed to find the optimal condition for processing broccoli, and its initial solution was obtained from the RSM. The results demonstrate a good performance of the GA for the optimization of the broccoli cooling process. The best conditions of vacuum cooling process were as follows: a weight between 273.5 g and 278.0 g, a water volume of 3.0% v/v, a processing time of 40 min, a pressure of 200 Pa, and a weight loss and end temperature of 0.34 ± 0.01% and 2.0 ± 0.0 °C, respectively, leading to a percentage of profit of 99.66 ± 0.01%.

Artificial intelligence based system to improve the inspection of plastic mould surfaces

Plastic industry is today in a constant growth, demanding several products from other segments, which includes the plastic moulds, used mainly in the injection moulding process. This paper presents a methodology for the surface evaluation of plastic moulds, aiming the automation of the polishing surface analysis. Provided that this type of analysis by traditional procedures can be slow and expensive, the development of automatic system could lead to considerable improvements regarding the speed and reliability of information. The starting point of the evaluation procedure is the image generated by the laser light scattered over the sample mould surface that could be captured and analysed by image processing and artificial intelligence techniques. The results showed that the proposed system is able to mapping and classifying several damages over the polished surface and could be an alternative to reduce efficiently the costs and the spending time in mould surface inspection tasks.

Segmentation of Retinal Blood Vessels Based on Ultimate Elongation Opening

This paper proposes a method for segmentation of retinal blood vessels based on ultimate attribute opening (UAO). The proposed approach analyzes the space of numerical residues generated by UAO in order to select the residues extracted from elongated regions by means of an elongation shape descriptor. Thus, the residues extracted are used to define the ultimate elongation opening. Experimental results, using the public datasets DRIVE and STARE show that the proposed approach is fast, simple and comparable to other methods found in the literature.

A Methodology for Sensory Evaluation of Food Products Using Self-Organizing Maps and K-Means Algorithm

Sensory analysis has an important impact on food production since its results can help the understanding of consumers’ perceptions about the products. Thus, many methods have been proposed and applied to quantify sensory attributes of food products. In this paper we proposed a methodology, using Kohonens Self-Organizing Maps and K-means algorithm, to classify food samples through the responses, provided by human evaluators, for their attributes such as aroma, flavor, appearance and texture. Conducted experiments in sensory analysis to determine the acceptance of new gelatins produced from chicken feet and new wines produced from spares of Açaí and Cajá confirm that proposed methodology is suitable for the investigated purpose.

A Fast and Robust Approach for Touching Grains Segmentation

The visual properties of agricultural grains are important factors for determining their market prices and assisting their choices by consumers. Despite the importance of visual inspection processes for agricultural grains quality, such tasks are usually handled manually and therefore subject to many failures. Thus, a computer vision approach that is able to segment correctly the grains contained in an image for further classification and detection of defects consists of an important practical application, which can be employed by visual quality inspection systems. In this work we propose an approach based on mathematical morphology and correlation-based granulometry techniques, guided by a set of heuristics, for grains segmentation. Experimental results showed that the proposed approach is able to segment the grains contained in an image, with high accuracy and very low computational time, even in cases where there are many grains glued together (touching grains).

An Intelligent Vision-Based System Applied to Visual Quality Inspection of Beans

In this work it is proposed an intelligent vision-based system for automatic classification of beans most consumed in Brazil. The system is able to classify the grains contained in a sample according to their skin colors, and is composed by three modules: image acquisition and pre-processing; segmentation of grains and classification of grains. In the conducted experiments, we used an apparatus controlled by a PC that includes a conveyor belt, an image acquisition chamber and a camera, to simulate an industrial line of production. The results obtained in the performed experiments indicate that the proposed system could be applied to visual quality inspection of beans produced in Brazil, since one of the steps in this process is the measurement of the mixture contained in a sample, taking into account the skin color of grains, for determining the predominant class of product and, consequently, its market price.

A Computer Vision System for Automatic Classification of Most Consumed Brazilian Beans

In this work we propose a computer vision system (CVS) for automatic classification of beans. It is able to classify the beans most consumed in Brazil, according to their skin colors and is composed by three main steps: (i) image acquisition and pre-processing, (ii) segmentation of grains and (iii) classification of grains. In the conducted experiments, we used an apparatus controlled by a PC that includes a conveyor belt, an image acquisition chamber and a camera, to simulate an industrial line of production. The results obtained in the experiments indicate that proposed system could be used to support the visual quality inspection of Brazilian beans.

A comparative study of statistical methods for characterisation of materials surfaces by means of texture analysis

Texture is an important attribute to distinguish objects and materials. Thus, along the decades many texture analysis methods have been proposed and utilised in a variety of application domains. Due to the fact there is not a generic method to describe a large variety of textures, comparative studies among the related methods became necessary. This paper describes a comparative study of the main statistical methods applied to materials surface characterisation. In order to evaluate the performance of the compared methods, an unsupervised neural network was used to classify a set of 3,000 textures images, divided in five categories, with different levels of details. Inferences from this work could assist those ones that intend to perform some tasks involving automatic inspection of texture, mainly in materials science context.

An Expert System for Improving Sieve Calibration Process

The reliability of the results obtained from instruments calibration is a problem frequently found in the calibration laboratories, especially when these instruments are mechanical and do not have a built-in communication interface. In this case, the time consuming is increased significantly and the calibration may be subject to human error. In this paper, a machine vision based system for automatic calibration of sieve was presented. The proposed equipment joined to the proposed technique showed in the results a reduction of 97% in the spending time for calibration process when compared to the traditional methods with the same accuracy.