No description available.

No description available.

No description available.

The focus of this research is to study how targets can be more faithfully detected in a multi-camera CCTV network system using spectral feature for the detection. The objective of the work is to develop colour constancy (CC) methodology to help maintain the spectral feature of the scene into a constant stable state irrespective of variable illuminations and camera calibration issues. Unlike previous work in the field of target detection, two versions of CC algorithms have been developed during the course of this work which are capable to maintain colour constancy for every image pixel in the scene: 1) a method termed as Enhanced Luminance Reflectance CC (ELRCC) which consists of a pixel-wise sigmoid function for an adaptive dynamic range compression, 2) Enhanced Target Detection and Recognition Colour Constancy (ETDCC) algorithm which employs a bidirectional pixel-wise non-linear transfer PWNLTF function, a centre-surround luminance enhancement and a Grey Edge white balancing routine. The effectiveness of target detections for all developed CC algorithms have been validated using multi-camera ‘Imagery Library for Intelligent Detection Systems’ (iLIDS), ‘Performance Evaluation of Tracking and Surveillance’ (PETS) and ‘Ground Truth Colour Chart’ (GTCC) datasets. It is shown that the developed CC algorithms have enhanced target detection efficiency by over 175% compared with that without CC enhancement. The contribution of this research has been one journal paper published in the Optical Engineering together with 3 conference papers in the subject of research.

No description available.

Doctor of Philosophy / This thesis investigates the role of colour in the cognitive processesing of emotional information. The research is guided by the effect of colour diagnosticity which has been shown previously to influence recognition performance of several types of objects as well as natural scenes. The research presented in Experiment 1 examined whether colour information is considered a diagnostic perceptual feature of seven emotional categories: happiness, sadness, anger, fear, disgust, surprise and neutral. Participants (N = 119), who were naïve to the specific purpose and expectations of the experiment, chose colour more than any other perceptual quality (e.g. shape and tactile information) as a feature that describes the seven emotional categories. The specific colour features given for the six basic emotions were consistently different from those given to the non-emotional neutral category. While emotional categories were often described by chromatic colour features (e.g. red, blue, orange) the neutral category was often ascribed achromatic colour features (e.g. white, grey, transparent) as the most symptomatic perceptual qualities for its description. The emotion 'anger' was unique in being the only emotion showing an agreement higher that 50% of the total given colour features for one particular colour - red. Confirming that colour is a diagnostic feature of emotions led to the examination of the effect of diagnostic colours of emotion on recognition memory for emotional words and faces: the effect, if any, of appropriate and inappropriate colours (matched with emotion) on the strength of memory for later recognition of faces and words (Experiments 2 & 3). The two experiments used retention intervals of 15 minutes and one week respectively and the colour-emotion associations were determined for each individual participant. Results showed that regardless of the subject’s consistency level in associating colours with emotions, and compared with the individual inappropriate or random colours, individual appropriate colours of emotions significantly enhance recognition memory for six basic emotional faces and words. This difference between the individual inappropriate colours or random colours and the individual appropriate colours of emotions was not found to be significant for non-emotional neutral stimuli. Post hoc findings from both experiments further show that appropriate colours of emotion are associated more consistently than inappropriate colours of emotions. This suggests that appropriate colour-emotion associations are unique both in their strength of association and in the form of their representation. Experiment 4 therefore aimed to investigate whether appropriate colour-emotion associations also trigger an implicit automatic cognitive system that allows faster naming times for appropriate versus inappropriate colours of emotional word carriers. Results from the combined Emotional-Semantic Stroop task confirm the above hypothesis and therefore imply that colour plays a substantial role not only in our conceptual representations of objects but also in our conceptual representations of basic emotions. The resemblance of the present findings collectively to those found previously for objects and natural scenes suggests a common cognitive mechanism for the processing of emotional diagnostic colours and the processing of diagnostic colours of objects or natural scenes. Overall, this thesis provides the foundation for many future directions of research in the area of colour and emotion as well as a few possible immediate practical implications.

Doctor of Philosophy (PhD) / The overall aim of this research was to examine aesthetic response to façade colour. Drawing on a range of theories and studies from environment-behaviour studies (EBS), Nasar’s (1994) probabilistic model of aesthetic response to building attributes provided a theoretical framework within which to examine patterns of response. Prompted by the Development Control Plan for Sydney Regional Environmental Plan: Sydney Harbour Catchment (NSWDOP, 2005), this research also linked its aims and methods to planning policy in Sydney. The main research questions focussed on whether changes in aesthetic response are associated with variations in façade colour; and whether changes in judgements about building size, congruity and preference are associated with differences in façade colour. A quasi-experimental research design was used to examine patterns of aesthetic response. The independent variable was represented by four façade colours in two classifications. An existing process, environmental colour mapping, was augmented with digital technology and used to isolate, identify and manipulate the independent variable and for preparation of visual stimuli (Foote, 1983; Iijima, 1995; Lenclos, 1977; Porter, 1997). Façade colour classifications were created from extant colour theories (including those of Albers, 1963; Hard & Sivik, 2001 and Itten, 1961). The façade colour classifications were further developed using F-sort and Q-sort methodology (Amin, 2000; Miller, Wiley & Wolfe, 1986; Stephenson, 1953). Ten dependent variables, linked to overall aesthetic response, were drawn from studies relating to environmental evaluation, building congruity and preference (Groat, 1992; Janssens, 2001; Russell, 1988; Russell, 2003; Russell, Ward & Pratt, 1981; Wohlwill & Harris, 1980). The dependent variables were presented in the form of a semantic differential rating scale and a sample group of 288 evaluated the visual stimuli. The Latin-square technique was used for the controlled presentation of visual stimuli. Factor analysis, correlation analysis and analysis of variance were applied to the data. The findings indicate that variations in aesthetic response are associated with differences in façade colour. Judgements about building size varied by up to 5% and buildings featuring contrasting façade colours were judged to be larger and more dominant. Judgements about a building’s congruity varied by up to 13% and buildings that featured harmonious colours were considered to be more congruous. Preference varied and harmonious façade colours were not necessarily preferred over contrasting façade colours. The outcomes from this research suggest that a new approach to façade colour within the context of planning policy may be appropriate. A model of façade colour evaluation is presented and, unlike current planning guidelines, the model allows for a participatory approach to façade colour evaluation and specification. The model allows for factors that may influence aesthetic response to façade colour (such as contextual, perceptual and idiographic factors) as well as variation in architectural expression with respect to façade colour.

Although most Holstein cattle are either black with white spots or red with white spots, two new coat colour traits called Variant Red (VR) and changeling, have recently been recognized by Holstein Canada. Variant red cattle look the same as red and white Holsteins but VR appears to be inherited as a dominant trait. Changeling cattle are born red and turn black by adulthood. Candidate genes for changeling included melanocortin 1 receptor (MC1R), agouti signalling protein (ASIP), attractin (ATRN), and melatonin receptor 1A (MTNR1A). Candidate genes for variant red included these same genes and β-defensin300 (putative K locus). Polymerase chain reaction restriction fragment length polymorphisms (PCR-RFLP) and sequencing were used to genotype single nucleotide polymorphisms (SNPs), which were recently discovered in our lab, for use in determining co-segregation in appropriate families. Cosegregation analysis excluded MC1R, ASIP, ATRN, or MTNR1A for changeling and VR, assuming single gene inheritance. Microsatellites and SNPs were used to map VR to a region on BTA27 surrounding and including β-defensin300 (LOD = 3.255). Cattle β-defensin300 was characterized from genomic DNA and mRNA from skin and appears to be homologous to canine β-defensin103 which is involved in black/red coat colour differentiation in many dog breeds. Although no mutations were found in the coding sequence, several were found in the 5, putative promoter region. No polymorphisms consistently and exclusively occurred in VR and not in changeling and/or black and white Holstein cattle. Common relatives occurred in the pedigrees of several variant red and changeling cattle. A hypothesis was developed which suggests that a common mutation is required for the occurrence of either phenotype. A dominant mutation in a second gene would be required for VR, potentially a promoter mutation in β-defensin300. Two recessive alleles at a yet unidentified second gene, potentially a promoter mutation in ASIP, would be required for the changeling phenotype to occur.

Although most Holstein cattle are either black with white spots or red with white spots, two new coat colour traits called Variant Red (VR) and changeling, have recently been recognized by Holstein Canada. Variant red cattle look the same as red and white Holsteins but VR appears to be inherited as a dominant trait. Changeling cattle are born red and turn black by adulthood. Candidate genes for changeling included melanocortin 1 receptor (MC1R), agouti signalling protein (ASIP), attractin (ATRN), and melatonin receptor 1A (MTNR1A). Candidate genes for variant red included these same genes and β-defensin300 (putative K locus). Polymerase chain reaction restriction fragment length polymorphisms (PCR-RFLP) and sequencing were used to genotype single nucleotide polymorphisms (SNPs), which were recently discovered in our lab, for use in determining co-segregation in appropriate families. Cosegregation analysis excluded MC1R, ASIP, ATRN, or MTNR1A for changeling and VR, assuming single gene inheritance. Microsatellites and SNPs were used to map VR to a region on BTA27 surrounding and including β-defensin300 (LOD = 3.255). Cattle β-defensin300 was characterized from genomic DNA and mRNA from skin and appears to be homologous to canine β-defensin103 which is involved in black/red coat colour differentiation in many dog breeds. Although no mutations were found in the coding sequence, several were found in the 5, putative promoter region. No polymorphisms consistently and exclusively occurred in VR and not in changeling and/or black and white Holstein cattle. Common relatives occurred in the pedigrees of several variant red and changeling cattle. A hypothesis was developed which suggests that a common mutation is required for the occurrence of either phenotype. A dominant mutation in a second gene would be required for VR, potentially a promoter mutation in β-defensin300. Two recessive alleles at a yet unidentified second gene, potentially a promoter mutation in ASIP, would be required for the changeling phenotype to occur.

Thesis--Ph. D., University of Hong Kong. / Typewritten.