Snake Validation: A PCA-Based Outlier Detection Method

Abstract

Abstract-We utilize outlier detection by principal component analysis (PCA) as an effective step to automate snakes/active contours for object detection. The principle of our approach is straightforward: we allow snakes to evolve on a given image and classify them into desired object and non-object classes. To perform the classification, an annular image band around a snake is formed. The annular band is considered as a pattern image for PCA. Extensive experiments have been carried out on oil-sand and leukocyte images and the performance of the proposed method has been compared with two other automatic initialization and two gradient-based outlier detection techniques. Results show that the proposed algorithm improves the performance of automatic initialization techniques and validates snakes more accurately than other outlier detection methods, even when considerable object localization error is present.

Automating Snakes for Multiple Objects Detection

“…3. Table 1 illustrates the number of seeds generated by the proposed QT, CoD and blind initialization (BI) [11]. CoD refers to the local maxima of the external Gradient Vector Flow (GVF) field.…”

“…Snake algorithms consist of three sequential steps: snake initialization, snake evolution and snake validation. For multiple objects detection, seeds are chosen inside the objects at the initialization step, then snakes are evolved from those seed points, and finally the evolved snakes are passed through a validation procedure to examine whether the snakes delineate the desired objects [11]. Substantial endeavors have taken place on the initialization and evolution steps towards snake automation.…”

“…Saha et al [11] proposed a snake validation scheme using principal component analysis (PCA). Their method places seeds blindly on the entire image and evolve one snake from each seed.…”

“…We propose a novel loss function that imposes smaller penalties on negative margins, and thus make Adaboost more robust to outliers. We exploit the advantages of multiple features including region, edge and shape over PCA-based intensity feature proposed earlier [11]. Our proposed initialization and validation can be successfully used as "plugins" with any existing snake evolution techniques.…”

Segment of Multiple Objects Based on Parameter Active Contour Model

“…In the second phase, segmentation of multiple objects in the image is executed by GVF. In [8][9], a seed point corresponds to one object, and this is the initial contour at the start of evolution. Then these seed points evolve into contours that contain objects under the expansion strategy.…”