Wood inspection with non-supervised clustering

Abstract. The appearance of sawn timber has huge natural variations that the human inspector easily compensates for mentally when determining the types of defects and the grade of each board. However, for automatic wood inspection systems these variations are a major source for complication. This makes it difficult to use textbook methodologies for visual inspection. These methodologies generally aim at systems that are trained in a supervised manner with samples of defects and good material, but selecting and labeling the samples is an error-prone process that limits the accuracy that can be achieved. We present a non-supervised clustering-based approach for detecting and recognizing defects in lumber boards. A key idea is to employ a self-organizing map (SOM) for discriminating between sound wood and defects. Human involvement needed for training is minimal. The approach has been tested with color images of lumber boards, and the achieved false detection and error escape rates are low. The approach also provides a self-intuitive visual user interface. Received: 16 December 2000 / Accepted: 8 December 2001 Correspondence to: O. Silvén     

一种改进的基于二元统计的HMM分词算法

中文分词是中文信息处理的基础。基于二元统计的HMM中文分词算法表现良好,但也存在易将包含常用介、副词的词进行误拆分的问题。改进的分词算法运用逆向最大匹配的思想,在计算粗分集权重的过程中,考虑了分词的词长及词序对正确切分的有利影响。该算法首先计算出二元统计粗分模型有向边的权值,然后根据词长修定权值,最后运用最短路径法求出分词结果。实验结果表明,该算法有效的解决了过分拆分的问题,分词效果良好。     

Cellular localization and changes in expression of prolactin receptor isoforms in sheep ovary throughout the estrous cycle

The actions of prolactin (PRL) on target cells depend on the type of prolactin receptor (PRLr) predominantly expressed, particularly whether the long PRLr isoform is expressed. The aims of this study were to determine the cellular localization and the changes in expression of long and short PRLr isoforms in sheep ovary throughout the estrous cycle. Long and short PRLrs were localized mostly in the same ovarian cells. Maximum signal intensity, particularly for long PRLrs, was found in stromal cells surrounding primordial and primary follicles, and, for both PRLrs, in granulosa cells of preantral follicles and in luteal cells. Moderate signal intensity for PRLrs was found in theca cells of preantral to ovulatory follicles, and in granulosa cells of antral follicles up to the gonadotropin-dependent stage. Decreasing immunoreactivity to PRLrs was found in granulosa cells of gonadotropin-dependent to ovulatory follicles. For long PRLrs in particular, no signal was found in mural granulosa cells of gonadotropin-dependent follicles; for both isoforms, no signal was found in most granulosa cells of ovulatory follicles. In primordial to gonadotropin-dependent follicles, cellular localization of PRLr was similar on days 0, 10 and 15 of the cycle. Oocytes consistently showed positive immunostaining for PRLrs. Comparative RT-PCR analysis of long and short PRLr expression showed that the short isoform is evenly expressed throughout the estrous cycle, whereas the expression of the long form increases at the time of estrus and decreases at mid-luteal phase and at the onset of the follicular phase. Expression of long PRLrs was greater than that of short PRLrs on day 0 of cycle; expression of both isoforms was similar on day 10 and on day 15, long PRLrs expression was lower than that of short PRLrs. Our results indicate that in sheep ovary, the maximum responsiveness to PRL might occur during the preovulatory phase of the estrous cycle.     

A genetic programming method for protein motif discovery and protein classification

Proteins can be grouped into families according to some features such as hydrophobicity, composition or structure, aiming to establish common biological functions. This paper presents MAHATMA—memetic algorithm-based highly adapted tool for motif ascertainment—a system that was conceived to discover features (particular sequences of amino acids, or motifs) that occur very often in proteins of a given family but rarely occur in proteins of other families. These features can be used for the classification of unknown proteins, that is, to predict their function by analyzing their primary structure. Experiments were done with a set of enzymes extracted from the Protein Data Bank. The heuristic method used was based on genetic programming using operators specially tailored for the target problem. The final performance was measured using sensitivity, specificity and hit rate. The best results obtained for the enzyme dataset suggest that the proposed evolutionary computation method is effective in finding predictive features (motifs) for protein classification.     

In vivo and in vitro antiinflammatory activity of saikosaponins

Buddlejasaponin I and saikosaponin 1 and 2, biologically active compounds from Scrophularia scorodonia and Bupleurum rigidum respectively, exert potent in vivo antiinflammatory effects on mouse ear edema induced by phorbol myristate acetate (PMA). The effects of these compounds on swelling and other inflammatory parameters are described. In screening for in vitro effects of saikosaponins on cellular systems generating cyclooxygenase (COX) and lipoxygenase (LOX) metabolites, we observed that most saikosaponins showed a significant effect. The action is more marked on LOX metabolite LTC4. Our data support the inhibition of arachidonic acid metabolism as one of the biochemical mechanisms that might be the rationale for the putative antiphlogistic activity of these saikosaponins.     

Role of transverse temperature gradients in the generation of lamellar eutectic solidification patterns

In directional solidification, a sample is pulled in an externally imposed temperature gradient. When the pulling direction and the temperature gradient are not perfectly aligned, a transverse temperature gradient is generated in the sample. We investigate the dynamics of lamellar eutectic patterns, both by experimental observations and by numerical simulations, and find that it is dramatically affected by a transverse gradient. Whereas, in its absence, the labyrinth patterns that are formed in the initial stage of solidification persist over the entire duration of the experiments, a transverse temperature gradient induces a drift of the pattern which rapidly generates a regular array of lamellae. The drift speed is found to be consistently larger than expected from the normal growth hypothesis, both in simulations and experiments.     

Anatomical basis for cell transplantation into mouse seminiferous tubules

Cell transplantation into the seminiferous tubules is a useful technique for the study of physiological and pathological conditions affecting the testis. However, the precise three-dimensional organization and, particularly, the complex connectivity of the seminiferous network have not yet been thoroughly characterized. To date, the technical approaches to address these issues have included manual dissection under the stereomicroscope, reconstruction of histological serial sections, and injection of contrast dyes, but all of them have yielded only partial information. Here, using an approach based on the microinjection of a self-polymerizing resin followed by chemical digestion of the surrounding soft tissues, we reveal fine details of the seminiferous tubule scaffold and its connections. These replicas of the testis seminiferous network were studied by scanning electron microscopy. The present results not only establish a morphological basis for more precise microinjection into the mouse seminiferous tubules but also enable a more profound investigation of physiological and embryological features of the testis.     

p-Cresol in cheese: Is it a flavouring compound or chemical contaminant?

ABSTRACT

p-Cresol has been identified as a flavouring compound in cheeses; however, scientific studies have already identified p-cresol as a potential chemical contaminant in environmental matrices. Thus, the objective of this study was to evaluate four traditional methods for extracting p-cresol from cheese samples in order to validate the best method, and finally to apply it to five cheese samples with different origins, processing and ripeness times. The analyses were performed by gas chromatography-mass spectrometry after derivatisation of p-cresol with anhydride acetic and pyridine. Better results were achieved by the QuEChERS method, which showed recovery higher than 80%, relative standard deviation lower than 16%, limit of quantification of 5 μg kg^?1 and linearity between 5 and 400 μg kg^?1 with R² 0.99. p-Cresol was quantified in almost all of the samples analysed at different concentration levels, which were in an increasing order at μg kg^?1: Cheddar (< LOQ), Parmesan (8 ± 0.7), Gorgonzola (103 ± 14), smoked Provolone (365 ± 28) and barbecue cheese (1001 ± 187). Although no maximum residue limit has been established for p-cresol in food, the results suggest that cheeses exposed to charcoal combustion notably increase the p-cresol levels and may represent a hazard to human health, especially in risk groups such as patients with chronic kidney disease who have serious problems with p-cresol.