Each year, the world’s top orange producers output millions of oranges for human consumption. This production is projected to grow by as much as 64 million in 2010 and so the demand for fast, low-cost and precise automated orange fruit grading systems is only deemed to become more increasingly important. There is however an underlying limit to most orange blemish detection algorithms. Most existing statistical-based, structural-based, model-based and transform-based orange blemish detection algorithms are plagued by the following problem: any pixels in an image of an orange having about the same magnitudes for the red, green and blue channels will almost always be classified as belonging to the same category (either a blemish or not). This however presents a big problem as the RGB components of the pixels corresponding to blemishes are very similar to pixels near the boundary of an orange. In light of this problem, this research utilizes a priori knowledge of the local intensity variations observed on rounded convex objects to classify the ambiguous pixels correctly. The algorithm has the effect of peeling-off layers of the orange skin according to gradations of the intensity. Therefore, any abrupt discontinuities detected along successive layers would significantly help identifying skin blemishes more accurately. A commercial-grade fruit inspection and distribution system was used to collect 170 navel orange images. Of these images, 100 were manually classified as good oranges by human inspection and the rest are blemished ones. We demonstrate the efficacy of the algorithm using these images as the benchmarking test set. Our results show that the system garnered 96% correctly classified good oranges and 97% correctly classified blemished oranges. The proposed system is easily customizable as it does not require any training. The fruit quality bands can be adjusted to meet the requirements set by the market standards by specifying an agreeable percentage of blemishes for each band.

Made available in DSpace on 2015-05-14T13:21:11Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 3039736 bytes, checksum: b5085a1b7abd18466b736a0a3c3e8de7 (MD5) Previous issue date: 2012-03-09 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This study proposes a digital image-based flow-batch analyzer (DIB-FBA) for the determination of Al(III) and Cr(VI) in waters. The proposed DIB-FBA uses a webcam with CCD sensor for acquisition of the images generated according to the RGB (Red-Green-Blue) colour system. The method for determination of Al(III) is based on reaction of this ion with quercetin and cetyltrimethylammonium bromide, yielding a yellow complex with maximum absorption at 428 nm. Chromium(VI) determination is accomplished by reaction of this ion with 1,5-diphenylcarbazide, whose product is a violet complex with maximum absorption at 548 nm. All calibration solutions were prepared in-line, and all analytical processes completed by simply changing the operational parameters in the DIB-FBA control software. The linear response ranges from 10 to 600 μg L−1 for Al (III) and from 10 to 300 μg L−1 for Cr (VI). Linear models for analytical curves were validated using ANOVA, F-test for lack of fit and residual plots. The limits of detection and sampling rates were estimated as 3,97 μg L−1 and 137 h−1, for Al(III) and 3,40 μg L−1 and 134 h−1, for Cr(VI), respectively. DIB-FBA performance was compared with the reference methods for determination of both analytes, and no statistically significant difference was observed between results by applying the paired t-test at a 95% confidence level. / Esse estudo propõe um analisador fluxo-batelada baseado em imagem digital (DIB-FBA) para determinação de Al(III) e Cr(VI) em águas. O DIB-FBA utiliza uma webcam com sensor CCD para a aquisição das imagens geradas de acordo com o sistema de cor RGB (Red-Green-Blue). O método para a determinação de Al(III) baseia-se na reação com quercetina e o brometo de cetiltrimetilamônio, produzindo um complexo amarelo com máximo de absorção em 428 nm. A determinação de cromo é baseada na reação do íon Cr(VI) com 1,5-difenilcarbazida, cujo o produto é um complexo de coloração violeta com máximo de absorção em 548 nm. Todas as soluções de calibração foram preparadas in-line e todos os processos analíticos foram realizados simplesmente mudando os parâmetros operacionais do aplicativo de controle do DIB-FBA. A resposta linear para o Al(III) se estende de 10 a 600 μg L−1 e para o Cr(VI) de 10 a 300 μg L−1. Os modelos lineares das curvas analíticas foram validados por intermédio de ANOVA, teste F para falta de ajuste e da análise gráfica dos resíduos deixados pelos modelos. Os limites de detecção e as frequências analíticas foram estimadas como sendo, respectivamente, 3,97 μg L−1 e 137 h−1, para o Al(III) e 3,40 μg L−1 e 134 h−1, para o Cr(VI). Os resultados da aplicação do DIB-FBA foram comparados com os obtidos empregando os respectivos métodos de referência. Constatou-se que não existe nenhuma diferença estatisticamente significativa entre os resultados, aplicando o teste t-pareado ao nível de confiança de 95%.

Made available in DSpace on 2015-05-14T13:21:45Z (GMT). No. of bitstreams: 1 parte1.pdf: 2505729 bytes, checksum: e872a92484e063cf1f264ba7b16d9e2e (MD5) Previous issue date: 2010-03-19 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This work proposes the use of digital images captured by a Webcam for determination of total acidity in red wines by means of an acid-base titration without using an external indicator. Digital images present the colour of the emergent radiation, which is complementary to the radiation absorbed by molecules of anthocyanines presents in wines. Each image, generated according to RGB system, yields a matrix of the values of the R, G and B components whose averages define a the colour value obtained as: 2R ⋅2G ⋅2B. This value was adopted as analytical response to build titrations curves based on digital images (DIB). For a more precise localization of the end point, titrations curves were generated on the basis of the second derivative values of the analytical response. Anthocyanines present different colours according to medium pH and, from the variation of the colour values of the images obtained during the titration of the red wines, the end point could be detected with precision. The official method recommends the use of potenciometric titration for determination of total acidity in red wines. This method requires a dilution of the sample before the titration that must be carried out until a fixed value of pH (8.2 8.4). In order to illustrate the feasibility of the proposed method titrations involving the determination of total acidity in ten red wines samples were carried out. Results were compared to the ones obtained by potenciometric titration used as reference method. No statistic difference has been observed between the results by applying the paired t-test at 95% confidence level. The proposed method yielded results with a better precision than the official method. These advantageous characteristics are attributed to the trivariate nature of the measures associated to digital images. / Neste trabalho, propõe-se o uso de imagens digitais, capturadas com uma Webcam, para determinação da acidez total de vinhos tintos por meio de titulação ácido-base sem a utilização de indicador externo. As imagens digitais apresentam a cor da radiação emergente que é complementar à da radiação absorvida por moléculas de antocianinas presentes nos vinhos. Cada imagem, gerada de acordo com o sistema vermelho-verde-azul (RGB), fornece uma matriz de valores dos componentes R, G e B cujas médias definem o valor de cor obtido como: 2R ⋅2G ⋅2B. Esse valor foi adotado como resposta analítica para a construção das curvas de titulação baseadas em imagens digitais (DIB). Para a localização mais precisa do ponto final, foram geradas as curvas de titulação com base nos valores da segunda derivada da resposta analítica. As antocianinas apresentam diferentes cores de acordo com o pH do meio e, a partir das variações do valor de cor das imagens obtidas durante a titulação dos vinhos tintos, o ponto final pôde ser detectado com precisão. O método oficial recomenda o uso da titulação potenciométrica para a determinação da acidez total de vinhos tintos. Este método requer diluição da amostra antes da titulação que deve ser efetuada até um valor fixo de pH (8,2-8,4). A fim de ilustrar a viabilidade do método proposto, foram realizadas titulações envolvendo a determinação da acidez total em dez amostras de vinhos tintos. Os resultados foram comparados com os obtidos pela titulação potenciométrica usada como método de referência. Constatou-se que não há diferença sistemática estatisticamente significativa entre os resultados, aplicando-se o teste t emparelhado ao nível de 95% de confiança. O método proposto produziu resultados com uma precisão melhor que a do método oficial. Essas características vantajosas da titulação DIB são atribuídas à natureza trivariada das medidas associadas às imagens digitais.

Made available in DSpace on 2015-05-14T13:21:30Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 12264651 bytes, checksum: c1dad0a27d50ddb05acdcc8e21042b61 (MD5) Previous issue date: 2012-09-11 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / In this work the potential of the Digital Image-Based Flame Emission Spectrometry (DIB-FES) is demonstrated through two completely separate applications. The first one consists in the indirect determination of three drugs in injectable form: sodium dicofenac, sodium dipyrone and clacium gluconate and the second in the combination of DIB-FES with Multiple Linear Regression (MLR) for the simultaneous determination of sodium and calcium in powder milk. Up to the present moment the literature does not report the use of traditional FES in: indirect determination of organic substances, simultaneous determination of analytes using a unique detector and overcoming the problem of spectral interference. In DIB-FES digital images of the flame are captured by webcam in its oxidant region (2,5 cm over the burner of the flame photometer) and are associated with the radiation emitted by metals present in the air-butane flame. Based on Red-Green-Blue (RGB) colour system, univariate and multivariate calibration models were developed, which were validated and then applied to real samples. In each application the results were compared with the results obtained by their respective reference methods. There were no statistically significant differences between the results when the paired t-test at the 95% confidence level. The estimated precision was better than the respective reference methods and accuracy was assessed by high values of bias and recovery rates between 97 and 104% in the two applications. / Neste trabalho a potencialidade da Espectrometria de Emissão em Chama Baseada em Imagens digitais (DIB-FES) é demonstrada por meio de duas aplicações completamente distintas. A primeira consiste na determinação indireta de três fármacos em soluções injetáveis: diclofenaco sódico, dipirona sódica e gluconato de cálcio e a segunda na combinação da DIB-FES com Regressão Linear Múltipla (MLR) para a determinação simultânea de sódio e cálcio em amostras de leite em pó. Até o presente momento a literatura não reporta a utilização da FES tradicional na: determinação indireta de substâncias orgânicas, determinação simultânea de analitos utilizando um único detector e superação do problema de interferência espectral. Na DIB-FES as imagens digitais da chama são capturadas pela webcam na região oxidante da mesma (2,5 cm acima do queimador do fotômetro) e estão associadas às radiações emitidas pelos metais presentes na chama ar-butano. Com base no sistema de cores Vermelho-Verde-Azul (RGB) foram desenvolvidos modelos de calibração univariada e multivariada, os quais foram validados e então aplicados em amostras reais. Em cada aplicação os resultados obtidos foram comparados com os resultados obtidos por seus respectivos métodos de referência. Não houve diferenças estatisticamente significativas entre resultados ao aplicar o teste t emparelhado ao nível de 95% de confiança. A precisão estimada foi melhor do que a de seus respectivos métodos de referência e a exatidão foi verificada por altas taxas de tendência e fatores de recuperação entre 97 e 104% nas duas aplicações.

This thesis deals with an electronic circuit and a PCB of a modular RGB LED display with the Ehernet interface. Firstly, author describes a RGB colour model, features of RGB LED displays, ways of control them. The next chapter contains a short description of the Ethernet interface, UDP and TCP protocols and a lwIP TCP/IP stack. The last theoretical chapter is an introduction to ARM Cortex-M3 and Cortex-M4 based microcontrollers. The next chaper is deals with a hardware design of the modular RGB LED display. The device is designed to be modular. Individual devices can be combosed together and create a larger display. Data from the Ethernet interface will be displayed on the RGB LED matrix, resolution of the matrix is 32 × 32 (1024 diodes). A refresh frequency is 100 Hz, a color depth is High color (16 bits) and a scanning 1/16 (two rows is driven at the same time). The next chapter describes the firmware for the RGB LED display, all its logical parts including a web page. Author also created the PC application, which sends pictures using UDP protocol to individual modules.

This thesis proposes a novel approach to extracting pose information from image sequences. Current state of the art techniques focus exclusively on the image space occupied by the body for pose and action recognition. The method proposed here, however, focuses on the negative spaces: the areas surrounding the individual. This has resulted in the colour-coded negative space approach, an image preprocessing step that circumvents the need for complicated model fitting or template matching methods. The approach can be described as follows: negative spaces surrounding the human silhouette are extracted using horizontal and vertical scanning processes. These negative space areas are more numerous, and undergo more radical changes in shape than the single area occupied by the figure of the person performing an action. The colour-coded negative space representation is formed using the four binary images produced by the scanning processes. Features are then extracted from the colour-coded images. These are based on the percentage of area occupied by distinct coloured regions as well as the bounding box proportions. Pose clusters are identified using feedback from an independent action set. Subsequent images are classified using a simple Euclidean distance measure. An image sequence is thus temporally segmented into its corresponding pose representations. Action recognition simply becomes the detection of a temporally ordered sequence of poses that characterises the action. The method is purely vision-based, utilising monocular images with no need for body markers or special clothing. Two datasets were constructed using several actors performing different poses and actions. Some of these actions included actors waving their arms, sitting down or kicking a leg. These actions were recorded against a monochrome background to simplify the segmentation of the actors from the background. The actions were then recorded on DV cam and digitised into a data base. The silhouette images from these actions were isolated and placed in a frame or bounding box. The next step was to highlight the negative spaces using a directional scanning method. This scanning method colour-codes the negative spaces of each action. What became immediately apparent is that very distinctive colour patterns formed for different actions. To emphasise the action, different colours were allocated to negative spaces surrounding the image. For example, the space between the legs of an actor standing in a T - pose with legs apart would be allocated yellow, while the space below the arms were allocated different shades of green. The space surrounding the head would be different shades of purple. During an action when the actor moves one leg up in a kicking fashion, the yellow colour would increase. Inversely, when the actor closes his legs and puts them together, the yellow colour filling the negative space would decrease substantially. What also became apparent is that these coloured negative spaces are interdependent and that they influence each other during the course of an action. For example, when an actor lifts one of his legs, increasing the yellow-coded negative space, the green space between that leg and the arm decreases. This interrelationship between colours hold true for all poses and actions as presented in this thesis. In terms of pose recognition, it is significant that these colour coded negative spaces and the way the change during an action or a movement are substantial and instantly recognisable. Compare for example, looking at someone lifting an arm as opposed to seeing a vast negative space changing shape. In a controlled research environment, several actors were instructed to perform a number of different actions. After colour coding the negative spaces, it became apparent that every action can be recognised by a unique colour coded pattern. The challenge is to ascribe a numerical presentation, a mathematical quotation, to extract the essence of what is so visually apparent. The essence of pose recognition and it's measurability lies in the relationship between the colours in these negative spaces and how they impact on each other during a pose or an action. The simplest way of measuring this relationship is by calculating the percentage of each colour present during an action. These calculated percentages become the basis of pose and action recognition. By plotting these percentages on a graph confirms that the essence of these different actions and poses can in fact been captured and recognised. Despite variations in these traces caused by time differences, personal appearance and mannerisms, what emerged is a clear recognisable pattern that can be married to an action or different parts of an action. 7 Actors might lift their left leg, some slightly higher than others, some slower than others and these variations in terms of colour percentages would be recorded as a trace, but there would be very specific stages during the action where the traces would correspond, making the action recognisable.In conclusion, using negative space as a tool in human pose and tracking recognition presents an exiting research avenue because it is influenced less by variations such as difference in personal appearance and changes in the angle of observation. This approach is also simplistic and does not rely on complicated models and templates