Repository logo
 
Loading...
Profile Picture

Search Results

Now showing 1 - 10 of 66
  • Region and graph-based motion segmentation
    Publication . Monteiro, Fernando C.; Campilho, Aurélio
    This paper describes an approach for integrating motion estimation and region clustering techniques with the purpose of obtaining precise multiple motion segmentation. Motivated by the good results obtained in static segmentation we propose a hybrid approach where motion segmentation is achieved within a region-based clustering approach taken the initial result of a spatial pre-segmentation and extended to include motion information. Motion vectors are first estimated with a multiscale variational method applied directly over the input images and then refined by incorporating segmentation results into a region-based warping scheme. The complete algorithm facilitates obtaining spatially continuous segmentation maps which are closely related to actual object boundaries. A comparative study is made with some of the best known motion segmentation algorithms.
  • Erythrocytes tracking through a microchannel using an automatic method of image analysis
    Publication . Taboada, Bruna Rafaela Pereira; Monteiro, Fernando C.; Lima, Rui A.
    In this study, an image processing and analyzing method has been developed in the MATLAB environment, to characterize the motion of RBCs flowing through a microchannel with a smooth contraction shape.
  • Interactive/automated method to count bacterial colonies
    Publication . Monteiro, Fernando C.; Ribeiro, J.E.; Martins, Ramiro
    The number of colonies in a culture is counted to calculate the concentration of bacteria in the original broth; however, manual counting can be tedious, time-consuming and imprecise. Automation of colony counting has been of increasing interest for many decades, and these methods have been shown to be more consistent than manual counting. Significant limitations of many algorithms used in automated systems are their inability to recognize overlapping colonies as distinct and to count colonies on the plate boundary. This study proposes an interactive semi-automated counting system and a fully automated counting system using image processing methods which overcomes these problems. The proposed systems are capable to reduce the manpower and time required for counting colonies while taking account colonies both around the central area and boundary areas of a Petri dish.
  • Overview of robotic based system for rehabilitation and healthcare
    Publication . Chellal, Arezki Abderrahim; Lima, José; Fernandes, Florbela P.; Gonçalves, José; Pacheco, Maria F.; Monteiro, Fernando C.
    As in many other fields, robots are increasingly being used in the healthcare sector, particularly for hospital logistics support, surgery and rehabilitation. Rehabilitation is a concern for millions of people around the world, and because of this, there has been a constant progress over the last decade in the rehabilitation robotics field, with the use of new technologies aimed at overcoming the different challenges faced in this field. In this sense, this paper reviews the main applications developed in the last ten years of rehabilitation robotics, as well as the different challenges that still need to be addressed in order to achieve the design of a prototype that is easy to use, small, safe, less costly and brings real added value to this field. Much of the efforts of the researchers in this topics is focused on providing as many DOF and ROM as possible, and also on the designing of new robots control algorithms
  • Performance analysis of speckle ultrasound image filtering
    Publication . Rosa, Rui Gonçalo Silva; Monteiro, Fernando C.
    Over the last three decades, several despeckling filters have been developed by researchers to reduce the speckle noise inherently present in ultrasound B-scan images without losing the diagnostic information. This paper compiles and compares well-known techniques mostly used in the smoothing or suppression of speckle noise in ultrasound images. A comparison of the methods studied is done based on an experiment, using quality metrics, texture analysis and interpretation of profiles to evaluate their performance and show the benefits each one can contribute to denoising and feature preservation. To test the methods, a noise-free image of a kidney is used and then the Field II program simulates a B-mode ultrasound image. By this way, the smoothing techniques can be compared using numeric metrics, taking the noise-free image as a reference. In this study, a total of 17 different speckle reduction algorithms have been documented based on spatial filtering, diffusion filtering and wavelet filtering, with 15 qualitative metrics estimation. We use the tendencies observed in our study in real images. A new evaluation metric is proposed to evaluate the despeckling results.
  • The role of background colour in pollen recognition task using CNN
    Publication . Monteiro, Fernando C.; Pinto, Cristina M.; Rufino, José
    Pollen recognition is a crucial but challenging task in addressing a variety of questions like pollination or palaeobotany, but also for other fields of research, e.g., allergology, melissopalynology or forensics. State-of-the-art methods mainly use deep learning CNNs for pollen recognition, however, we observe that existing published approaches use original images without study the possible biased recognition due to pollen’s background colour. In this paper, we evaluate the DenseNet model trained with original images and with segmented images (remove background) and analyse network’s predictive performance under these conditions using a cross evaluation approach. An accuracy of 97.4% was achieved that represents one of the best successes rate when weighted with the number of taxa of any attempt at automated pollen analysis currently documented in the literature. From these results, we confirm the existence of background specific influence in the recognition task.
  • Tracking red blood cells flowing through a microchannel with a hyperbolic contraction: an automatic method
    Publication . Taboada, Bruna Rafaela Pereira; Monteiro, Fernando C.; Lima, Rui A.
    The present chapter aims to assess the motion and deformation index of red blood cells (RBCs) flowing through a microchannel with a hyperbolic contraction using an image analysis based method. For this purpose, a microchannel containing a hyperbolic contraction was fabricated in polydimethylsiloxane by using a soft-lithography technique and the images were captured by a standard high-speed microscopy system. An automatic image processing and analyzing method has been developed in a MATLAB environment, not only to track both healthy and exposed RBCs motion but also to measure the deformation index along the microchannel. The keyhole model has proved to be a promising technique to track automatically healthy and exposed RBCs flowing in this kind of microchannels.
  • Diagnostic performance of thermograms and mammograms for breast cancer detection
    Publication . Amaral, Elizabete Marques; Monteiro, Fernando C.
    In this paper we carry out a comparative study between mammograms and thermograms, involving 200 women, in order to bring a new light on the current situation of thermography as an effective tool to support mammography in prevention and monitoring the breast cancer.
  • Is diabetic retinopathy grading biased by imbalanced datasets?
    Publication . Monteiro, Fernando C.; Rufino, José
    Diabetic retinopathy (DR) is one of the most severe complications of diabetes and the leading cause of vision loss and even blindness. Retinal screening contributes to early detection and treatment of diabetic retinopathy. This eye disease has five stages, namely normal, mild, moderate, severe and proliferative diabetic retinopathy. Usually, highly trained ophthalmologists are capable of manually identifying the presence or absence of retinopathy in retinal images. Several automated deep learning (DL) based approaches have been proposed and they have been proven to be a powerful tool for DR detection and classification. However, these approaches are usually biased by the cardinality of each grade set, as the overall accuracy benefits the largest sets in detriment of smaller ones. In this paper, we applied several state-of-the-art DL approaches, using a 5-fold cross-validation technique. The experiments were conducted on a balanced DDR dataset containing 31330 retina fundus images by completing the small grade sets with samples from other well known datasets. This balanced dataset increases robustness of training and testing tasks as they used samples from several origins and obtained with different equipment. The results confirm the bias introduced by using imbalanced datasets in automatic diabetic retinopathy grading.
  • Interactive/automated method to count bacterial colonies
    Publication . Ribeiro, J.E.; Martins, Ramiro; Monteiro, Fernando C.
    The growth and maintenance of bacteria on agar plates (Petri dishes) has long been a common practice in microbiology. The number of colonies in a culture is usually counted manually to calculate the concentration of bacteria, however, this process is time-consuming, tedious and error prone. Most automated counting systems, existing on the literature, perform adequately when the colonies are well spaced, large, circular in shape and with good contrast from the background. When these assumptions are violated, most automated colony analysis systems can rapidly lose reliability, accuracy and utility. To address the above problems, the goal of this study is to design and implement a cost-effective, software-centred system that accepts general digital camera images as its input, for detecting as well as enumerating bacterial colonies in a fully automatic manner. An interactive semi-automatic system is also proposed to overcome any error from fully automatic system. The two systems (automated and interactive) combine thresholding, median filter and morphological operations to segment the colonies on a Petri dish image. The next step consists in separating the individual colonies from the clustered colonies. Then, the units colonies are counted. To separate and count the clustered colonies, the automatic system uses a watershed transformation and the interactive system uses the clicks of the user. The proposed systems are capable to reduce the manpower and time required for counting colonies while producing correct colony counting.