Victor Ayala-Ramirez

Circle detection on images using genetic algorithms

In this paper, we present a circle detection method based on genetic algorithms. Our genetic algorithm uses the encoding of three edge points as the chromosome of candidate circles (x,y,r) in the edge image of the scene. Fitness function evaluates if these candidate circles are really present in the edge image. Our encoding scheme reduces the search space by avoiding trying unfeasible individuals, this results in a fast circle detector. Our approach detects circles with sub-pixellic accuracy on synthetic images. Our method can also detect circles on natural images with sub-pixellic precision. Partially occluded circles can be located in both synthetic and natural images. Examples of the application of our method to the recognition of hand-drawn circles are also shown. Detection of several circles in a single image is also handled by our method.

Mobile robot path planning using artificial bee colony and evolutionary programming

Graphical abstractDisplay Omitted HighlightsWe solve the path planning problem using the combination of two evolutionary methods.First, an artificial bee colony (ABC) finds a feasible path in the free space.Second, evolutionary programming (EP) optimizes the path length and smoothness.The proposed approach was compared to a probabilistic roadmap (PRM) method.The ABC-EP approach outperforms the PRM approach on problems of varying complexity. In this paper, an evolutionary approach to solve the mobile robot path planning problem is proposed. The proposed approach combines the artificial bee colony algorithm as a local search procedure and the evolutionary programming algorithm to refine the feasible path found by a set of local procedures. The proposed method is compared to a classical probabilistic roadmap method (PRM) with respect to their planning performances on a set of benchmark problems and it exhibits a better performance. Criteria used to measure planning effectiveness include the path length, the smoothness of planned paths, the computation time and the success rate in planning. Experiments to demonstrate the statistical significance of the improvements achieved by the proposed method are also shown.

Path planning using genetic algorithms for mini-robotic tasks

We present a genetic algorithm-based method to optimize trajectory planning for mini-robotic tasks. Codifying a number of motion primitive parameters into computational chromosomes does this. Each trajectory is composed of a fixed number N of straight segments. We search with a genetic algorithm the length and direction parameters of the N path segments that let us to arrive a target position from the current robot position. We show design choices of the genetic operators (selection, mutation and fitness function) used in our genetic algorithm implementation. We present simulations of our method and experimentation on a mini-robotic platform is implemented.

Active tracking based on Hausdorff matching

We present an object tracking system based on an edge model for the target characterization. The target position is estimated by looking for the model in the current image using a Hausdorff partial distance. Target is searched only in a sub-window of current edge image. Its boundaries are determined by Kalman filter estimation that uses target dynamics to predict the current position. We use a spiral searching strategy to find the actual position. The target model is updated in each iteration by using unidirectional partial distance from the image to the model. This model is refined by an enclosure operator in order to perform the target/background discrimination. The parameters of our system can be modified in an active way along the tracking task. The system is shown to be robust to illumination changes and pose variations. The system has been also embedded in a mobile robot for personal robotics applications and integrated in a real-time OS (3 Hz).

Bin-packing using genetic algorithms

We present in this paper a genetic algorithm (GA) approach to solve 2D bin packing problems of polygonal shapes on a rectangular canvas. We present the way to encode shape parameters and a fitness function based on a the medial axis transform (MAT) to evaluate individuals of a genetic algorithm population. Some test and results of our experimentation are presented.

Fast texel size estimation in visual texture using homogeneity cues

In this paper, we address the texel size estimation of periodic and near-periodic texture images. Such a problem has shown to be difficult when corrupted and distorted patterns are analyzed, and the accuracy and robustness are significant. In this study, we propose the use of the homogeneity cues computed using a difference histogram. Varying the displacement vector that relates two pixels and localizing the resultant homogeneity maximum value, we can automatically determine the texel size. Experiments were carried out in order to evaluate the performance of our method. Results on artificially distorted images and on natural near-periodic images, show that the proposed approach is more accurate and robust than other state-of-the-art methods. Furthermore, the computation of homogeneity cues is not intricate nor time-consuming, and hence, it can be considered for practical applications where computation time is critical.

Comparison of perceptual color spaces for natural image segmentation tasks

Color image segmentation largely depends on the color space chosen. Furthermore, spaces that show perceptual uniformity seem to outperform others due to their emulation of the human perception of color. We evaluate three perceptual color spaces, CIELAB, CIELUV, and RLAB, in order to determine their contribution to natural image segmentation and to identify the space that obtains the best results over a test set of images. The nonperceptual color space RGB is also included for reference purposes. In order to quantify the quality of resulting segmentations, an empirical discrepancy evaluation methodology is discussed. The Berkeley Segmentation Dataset and Benchmark is used in test series, and two approaches are taken to perform the experiments: supervised pixelwise classification using reference colors, and unsupervised clustering using k-means. A majority filter is used as a postprocessing stage, in order to determine its contribution to the result. Furthermore, a comparison of elapsed times taken by the required transformations is included. The main finding of our study is that the CIELUV color space outperforms the other color spaces in both discriminatory performance and computational speed, for the average case.

Fuzzy color tracking for robotic tasks

We present in this paper how to track an object by using a color cue. Color is represented as fuzzy membership functions in the CIELab color space. Integration of the fuzzy representation and CIELab color space make our tracking system robust to illumination variability. Target initialization is done in an interactive way, user selects a color target and membership functions for each coordinates are then defined. Target search is done by examining pixel intensities over a test region in the current image using a fuzzy logic rule. We have tested our approach experimentally and our system can track a colored object at rates of about 15Hz on a Pentium computer. A visual servoing system that uses our target tracking system for feature extraction has also been developed.

Real-Time Hand Posture Recognition for Human-Robot Interaction Tasks

In this work, we present a multiclass hand posture classifier useful for human-robot interaction tasks. The proposed system is based exclusively on visual sensors, and it achieves a real-time performance, whilst detecting and recognizing an alphabet of four hand postures. The proposed approach is based on the real-time deformable detector, a boosting trained classifier. We describe a methodology to design the ensemble of real-time deformable detectors (one for each hand posture that can be classified). Given the lack of standard procedures for performance evaluation, we also propose the use of full image evaluation for this purpose. Such an evaluation methodology provides us with a more realistic estimation of the performance of the method. We have measured the performance of the proposed system and compared it to the one obtained by using only the sampled window approach. We present detailed results of such tests using a benchmark dataset. Our results show that the system can operate in real time at about a 10-fps frame rate.

Natural Image Segmentation Using the CIELab Space

In this work, we have tested a color based segmentation approach that uses the CIELab color space. We use a color reduction approach where the dominant color database is obtained from the analysis of the entire Berkeley natural image database. After experiments using single color components and all their possible combinations as the segmentation basis, we have found that ab is the best color component combination for the segmentation task. F measures and Precision Recall graphs are used as the evidence for this conclusion.