Browsing by Author "Soto Arriaza, Álvaro"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Automated Detection of Fish Bones in Salmon Fillets Using X-ray Testing
    (IEEE, 2010) Mery Quiroz, Domingo Arturo; Lillo Valles, Iván Alberto; Löbel Díaz, Hans-Albert; Riffo Bouffanais, Vladimir; Soto Arriaza, Álvaro; Cipriano, Aldo; Aguilera Radic, José Miguel
    X-ray testing is playing an increasingly important role in food quality assurance. In the production of fish fillets, however, fish bone detection is performed by human operators using their sense of touch and vision which can lead to misclassification. In countries where fish is often consumed, fish bones are some of the most frequently ingested foreign bodies encountered in foods. Effective detection of fish bones in the quality control process would help avoid this problem. For this reason, we developed an X-ray machine vision approach to automatically detect fish bones in fish fillets. This paper describes our approach and the corresponding validation experiments with salmon fillets. The approach consists of six steps: 1) A digital X-ray image is taken of the fish fillet being tested. 2) The X-ray image is filtered and enhanced to facilitate the detection of fish bones. 3) Potential fish bones in the image are segmented using band pass filtering, thresholding and morphological techniques. 4) Intensity features of the enhanced X-ray image are extracted from small detection windows that are defined in those regions where potential fish bones were segmented. 5) A classifier is used to discriminate between 'bones' and 'no-bones' classes in the detection windows. 6) Finally, fish bones in the X-ray image are isolated using morphological operations applied on the corresponding segments classified as 'bones'. In the experiments we used a high resolution flat panel detector with the capacity to capture up to a 6 million pixel digital X-ray image. In the training phase, we analyzed 20 representative salmon fillets, 7700 detection windows (10×10 pixels) and 279 intensity features. Cross validation yielded a detection performance of 95% using a support vector machine classifier with only 24 selected features. We believe that the proposed approach opens new possibilities in the field of automated visual inspection of salmon and other similar fish.
  • Thumbnail Image
    Item
    Hierarchical Joint Max-Margin Learning of Mid and Top Level Representations for Visual Recognition
    (IEEE, 2013) Löbel Díaz, Hans-Albert; Vidal Navarro, René Esteban; Soto Arriaza, Álvaro
    Currently, Bag-of-Visual-Words (BoVW) and part-based methods are the most popular approaches for visual recognition. In both cases, a mid-level representation is built on top of low-level image descriptors and top-level classifiers use this mid-level representation to achieve visual recognition. While in current part-based approaches, mid- and top-level representations are usually jointly trained, this is not the usual case for BoVW schemes. A main reason for this is the complex data association problem related to the usual large dictionary size needed by BoVW approaches. As a further observation, typical solutions based on BoVW and part-based representations are usually limited to extensions of binary classification schemes, a strategy that ignores relevant correlations among classes. In this work we propose a novel hierarchical approach to visual recognition based on a BoVW scheme that jointly learns suitable mid- and top-level representations. Furthermore, using a max-margin learning framework, the proposed approach directly handles the multiclass case at both levels of abstraction. We test our proposed method using several popular benchmark datasets. As our main result, we demonstrate that, by coupling learning of mid- and top-level representations, the proposed approach fosters sharing of discriminative visual words among target classes, being able to achieve state-of-the-art recognition performance using far less visual words than previous approaches.
  • Thumbnail Image
    Item
    Visual Recognition to Access and Analyze People Density and Flow Patterns in Indoor Environments
    (IEEE, 2015) Ruz Ruz, Cristian Daniel; Pieringer Baeza, Christian Philip; Peralta Marquez, Billy Mark; Lillo Valles, Iván Alberto; Espinace Ronda, Pablo Andrés; Gonzalez, R.; Wendt González, Bruno Nicolás; Mery Quiroz, Domingo Arturo; Soto Arriaza, Álvaro
    This work describes our experience developing a system to access density and flow of people in large indoor spaces using a network of RGB cameras. The proposed system is based on a set of overlapped and calibrated cameras. This facilitates the use of geometric constraints that help to reduce visual ambiguities. These constraints are combined with classifiers based on visual appearance to produce an efficient and robust method to detect and track humans. In this work, we argue that flow and density of people are low level measurements that need to be complemented with suitable analytic tools to bridge semantic gaps and become useful information for a target application. Consequently, we also propose a set of analytic tools that help a human user to effectively take advantage of the measurements provided by the system. Finally, we report results that demonstrate the relevance of the proposed ideas.