| Image segmentation is the operation of partitioning an image into a collection of connected sets of pixels, the image segmentation plays a very important role in medical images processing. It is a base technique for the key target area extraction, a key part of the organization expressed quantitatively as well as the basis of a three-dimensional image reconstruction. For human neuroanatomist, segmentation of neuro-images is a trivial task but, unfortunately, it is very time consuming. Therefore, accurate algorithms for automatic neuronal segmentation are indispensable for large scale geometric reconstruction of densely interconnected neuronal tissue. The neuron images have the characteristics of medical image imaging and they have its own characteristics with respect to general medical image as well. Nevertheless, its structure is complexity, such as intricate topology, various cell interference inside, and noisy textures. In addition, the poor quality of the imaging tools causes the border lacking and fuzzy. Those problems make the automatic segmentation of neuronal electron microscopy images very difficult. Therefore, the accurate segmentation needs more distinctive features in detail naturally. In order to solve the problems above, this paper focuses on two topics as follows:(1) It is sensitive to noise and time consuming to implement segmentation only with pixel level features. Also, pixel level features could not describe local consistency characteristic of EM images. superpixelis a natural form of enforcing local consistency while respecting original image boundaries. Recent studies on superpixels showed that superpixels had superior performance to using rectangular image windows for localized image processing. Moreover, superpixel, as a visual primitive, performs high efficiency for classification. we study the superpixel based feature extraction as well as the selection of appropriate classifier, and implement the segmentation approach based upon superpixel and random forest. Experimental results show the efficiency of our method on the data set of ISBI2012EM image Segmentation Challenge. (2) The accurate segmentation needs more distinctive features in detail. the methods based on hand-designed feature extraction require elaborately designed features and cannot process raw images as well as it need a deep understanding for a specific problem, making it uneasy to adapt to other domain. In addition, the hand-designed features only capture low-level edge information and it is difficult to design features that effectively capture mid-level cues (e.g. edge intersections) or highlevel representation (e.g. object parts), which is very important for neuron images. Also in some other applications, one may not have this knowledge that can be used to develop feature extractors. What’s more, recent developments in machine learning, known as "Deep Learning", have shown how hierarchies of features can be learned directly from data and automatic extraction methods become a tendency in the image processing. Based on Convolutional Neural Network (CNN) and Random Forest classifier (RF), a hybrid CNN-RF method for EM neuron segmentation is presented. CNN as a feature extractor is trained firstly, and then well behaved features are learned with the trained feature extractor automatically. Finally, Random Forest classifier is trained on the learned features to perform neuron segmentation. Experiments have been conducted on the benchmarks for the ISBI2012EM Segmentation Challenge, and the proposed method achieves the effectiveness results... |
PDF Full Text Request