Sivalogeswaran Ratnasingam

Study of the photodetector characteristics of a camera for color constancy in natural scenes

An algorithm is described to extract two features that represent the chromaticity of a surface and that are independent of both the intensity and correlated color temperature of the daylight illuminating a scene. For mathematical convenience this algorithm is derived using the assumptions that each photodetector responds to a single wavelength and that the spectrum of the illumination source can be represented by a blackbody spectrum. Neither of these assumptions will be valid in a real application. A new method is proposed to determine the effect of violating these assumptions. The conclusion reached is that two features can be obtained that are effectively independent of the daylight illuminant if photodetectors with a spectral response whose full width at half maximum is 80 nm or less are used.

Chromaticity Space for Illuminant Invariant Recognition

In this paper, an algorithm is proposed to extract two illuminant invariant chromaticity features from three image sensor responses. The algorithm extracts these chromaticity features at pixel level and therefore can perform well in scenes illuminated with nonuniform illuminant. An approach is proposed to use the algorithm with cameras of unknown sensitivity. The algorithm was tested for separability of perceptually similar colors under the International Commission on Illumination standard illuminants and obtained a good performance. It was also tested for color-based object recognition by illuminating objects with typical indoor illuminants and obtained a better performance compared to other existing algorithms investigated in this paper. Finally, the algorithm was tested for skin detection invariant to illuminant, ethnic background and imaging device. In this investigation, daylight scenes under different weather conditions and scenes illuminated by typical indoor illuminants were used. The proposed algorithm gives a better skin detection performance compared to widely used standard color spaces. Based on the results presented, the proposed illuminant invariant chromaticity space can be used for machine vision applications including illuminant invariant color-based object recognition and skin detection.

Object recognition based on tactile form perception

In this paper a self organising map is proposed for object recognition based on tactile form perception. A robot hand with three fingers, with the same number of degrees of freedom as the human hand, is used for obtaining the required tactile measurements. Finger joint angles were recorded when the hand was grasping different objects, in three different orientations. A self organising map was used to categorise objects based on the measurements obtained using the hand. The proposed system learnt to recognise objects of different shape, as well as objects of the same shape but different size. To test the generalisation ability of the system, new objects (different from the training set) were applied and it was observed that the system learnt to categorise objects based on their shape and size. This paper also investigates the reliability of object recognition using tactile form perception with noisy measurements, the maximum number of different objects that can be recognised without significantly degrading the performance and the ability to recognise similar shaped objects with small differences in dimensions. Based on the test results presented, the system can recognise 89 % of 25 different objects. This promising performance suggests that tactile form perception can be reliably used for object recognition in robotic applications.

A spiking neural network for tactile form based object recognition

This paper proposes a biologically plausible system for object recognition based on tactile form perception. A spiking neural network, an encoding scheme for converting the input values into spike trains, a method for converting the output spike pattern into reliable features for object recognition and a training approach for the spiking neural network are proposed. Three separate spiking neural networks are used in this recognition system. Three features, based on the output firing pattern of the three networks, are projected onto a three dimensional space. Each class of objects forms a cluster in the three-dimensional feature space. During the training the firing threshold of the hidden layer is modified in such a way that the cluster formed by an object is small and does not overlap with neighbouring clusters. The system has been tested with a number of objects for recognition based on shape. In addition, the system has also been tested for the ability to recognise objects of the same shape but different size. The results show the proposed system gives good performance in recognising objects based on tactile form perception.

Optimum sensors for color constancy in scenes illuminated by daylight

The apparent color of an object within a scene depends on the spectrum of the light illuminating the object. However, recording an objects color independent of the illuminant spectrum is important in many machine vision applications. In this paper the performance of a blackbody-model-based color constancy algorithm that requires four sensors with different spectral responses is investigated under daylight illumination. In this investigation sensor noise was modeled as gaussian noise, and the responses were quantized using different numbers of bits. A projection-based algorithm whose output is invariant to illuminant is investigated to improve the results that are obtained. The performance of both of these algorithms is then improved by optimizing the spectral sensitivities of the four sensors using freely available CIE standard daylight spectra and a set of lightness-normalized Munsell reflectance data. With the optimized sensors the performance of both algorithms is shown to be comparable to the human visual system. However, results obtained with measured daylight spectra show that the standard daylights may not be sufficiently representative of measured daylight for optimization with the standard daylight to lead to a reliable set of optimum sensor characteristics.

A comparison of encoding schemes for haptic object recognition using a biologically plausible spiking neural network

In this paper a biologically inspired spiking neural network based haptic object recognition system is proposed. A number of different encoding schemes to convert the haptic measurements into spike train are proposed and investigated for haptic object recognition and compared with existing encoding schemes. The spiking neural network was trained using a supervised training approach that is based on the steepest descent algorithm. During the training, firing threshold of the hidden layer neurons were modified in such a way that the ability of the system in recognising different objects is maximised. A multiplexing scheme is used to convert the parallel spike train of the hidden layer into a serial stream. To convert the output spike train into a reliable feature that represents the shape of an object, moment of the spikes with respect to a reference time is calculated. A robot hand with three fingers that have the same number of degrees of freedom as the human fingers was used for testing the system. The hand was made to grasp different objects and the joint angles were recorded. These recorded angles were converted into spike train using different encoding schemes and applied as input to the network. Test results show that the performance of the system varies depending on the input encoding scheme and with the best encoring scheme the system can recognise 100% of 7 different objects.

An Algorithm to Determine the Chromaticity Under Non-uniform Illuminant

Colour based object recognition is a difficult problem because of the effect of scene illuminant and geometry on the captured image. In this paper the ability of an algorithm proposed by Finlayson and Drew [1] to separate similar colours is assessed. A new variant of this algorithm is then proposed that results in a slight improvement in performance. A significant performance improvement is achieved by optimising the characteristics of the sensors that are used to acquire the data for this algorithm. This optimisation process results in several combinations of sensors and associated data projections that have a comparable performance when required to distinguish between similar colours. Since this performance is comparable to that of the human visual system it is suggested that with the correct sensors this algorithm is capable of obtaining useful chromaticity information under varying illumination conditions.

Illuminant spectrum estimation at a pixel

In this paper, an algorithm is proposed to estimate the spectral power distribution of a light source at a pixel. The first step of the algorithm is forming a two-dimensional illuminant invariant chromaticity space. In estimating the illuminant spectrum, generalized inverse estimation and Wiener estimation methods were applied. The chromaticity space was divided into small grids and a weight matrix was used to estimate the illuminant spectrum illuminating the pixels that fall within a grid. The algorithm was tested using a different number of sensor responses to determine the optimum number of sensors for accurate colorimetric and spectral reproduction. To investigate the performance of the algorithm realistically, the responses were multiplied with Gaussian noise and then quantized to 10 bits. The algorithm was tested with standard and measured data. Based on the results presented, the algorithm can be used with six sensors to obtain a colorimetrically good estimate of the illuminant spectrum at a pixel.

Analysis of colour constancy algorithms using the knowledge of variation of correlated colour temperature of daylight with solar elevation

In this article, we present an investigation of possible improvement of the colour constant reflectance features that can be obtained from daylight illuminated scenes using pixel-level colour constancy algorithms (model-based algorithm: S Ratnasingam, S Collins, J. Opt. Soc. Am. A 27, 286–294 (2010) and Projection-based algorithm: GD Finlayson, MS Drew, IEEE ICCV, 2001, pp. 473–480). Based on the investigation we describe a method to improve the performance of the colour constancy algorithms using the correlation between the correlated colour temperature of measured daylight with the solar elevation and phase of the day (morning, midday and evening). From this observation, the data from 1 year are used to create a solar elevation and phase of day-dependant method of interpreting the information obtained the colour constancy algorithms. Test results show that using the proposed method with 40-dB signal-to-noise ratio the performance of the projection-based algorithm and model-based algorithm can be improved on average by 33.7 and 45.4%, respectively. More importantly, a larger improvement (85.9 and 113.7%) was obtained during the middle period of each day which is defined as when the solar elevation is larger than 20°.

Optimum Sensors for `Chromaticity' Constancy in the Pixel

In machine vision systems recording the colour of an object is crucial for applications such as skin detection while it will enhance applications including colour based recognition and image retrieval. Unfortunately, almost none of the existing colour constancy algorithms have been designed to deal with the high dynamic ranges that can occur in external, naturally illuminated scenes. One algorithm that can deal with these scenes has been proposed by Finlayson and Drew. In this paper a method of assessing the performance of this algorithm, and equivalent algorithms, are proposed. The performance of this algorithm is then significantly improved by optimising the spectral response of the sensors used to obtain the data required by algorithm. Since the resulting performance is comparable to that of the human visual system it appears that this algorithm is capable of obtaining useful chromaticity information under highly varying illumination conditions.