Craig S. Rose

Aggregated segmentation of fish from conveyor belt videos

Automation of fishery survey through the aid of visual analysis has received increasing attention. In this paper, a novel algorithm for the aggregated segmentation of fish images taken from conveyor belt videos is proposed. The watershed algorithm driven by an automatic marker generation scheme successfully separates clustered fish images without damaging their boundaries. A target selection based on appearance classification then rejects non-fish objects. By applying histogram backprojection and kernel density estimation, an innovative algorithm for combining object masks of one tracked fish from multiple frames into a refined single one is also proposed. Experimental results show that accurate fish segmentation from conveyor belt videos is achieved.

Chute based automated fish length measurement and water drop detection

Image processing and analysis techniques have drawn increasing attention since they enable a non-extractive and non-lethal approach to fisheries survey, such as fish size measurement, abundance prediction, catch estimation and compliance, species recognition and population counting. In this work, we present an innovative and effective method for measuring the chute-based fish length based on the morphological midline of the fish. The midline is generated through recursive morphological operations on the segmented fish mask. To conduct reliable measurement, even under harsh environment, we also propose a systematic method for detecting water drop on camera lens. The robust detection, which can be performed either in real-time or in offline processing, is based on a blur measure derived from the gradient of the image and the contour of fish.

Shrinking Encoding with Two-Level Codebook Learning for Fine-Grained Fish Recognition

Bag-of-features (BoF) shows a great power in representing images for image classification. Many codebook learning methods have been developed to find discriminative parts of images for fine-grained recognition. Built upon BoF framework, we propose a novel approach for finegrained fish recognition with two-level codebook learning by shrinking coding coefficients. In the framework, only the maximum-valued coefficient will be maintained in the local spatial region if followed by max pooling strategy. However, the maximum-valued coefficient may result from a local descriptor which is not discriminative among fine-grained classes, resulting in difficulty in classification. In this paper, a two-level codebook is learned to represent the importance between the local descriptor and each codeword in its corresponding k-nearest neighbors. A shrinkage function is also introduced to shrink unrelated coefficients after encoding. Our experimental results show that the proposed method achieves significant performance improvement for fine-grained fish recognition tasks.

Uncertainty sampling based active learning with diversity constraint by sparse selection

Uncertainty based active learning has been well studied for selecting informative samples to improve the performance of the classifier. One of the simplest strategy is that we always select samples with top largest uncertainties for a query. However, the selected samples may be very similar to each other, which results in little information added to update the classifier. In other words, we should avoid selecting similar samples for training the classifier. This paper addresses this problem by proposing a novel method using uncertainty based active learning algorithm with diversity constraint by sparse selection. First, uncertainty scores of unlabeled samples are obtained based on the previously trained support vector machine (SVM) classifiers. Then the sample selection is represented as a sparse modeling problem and optimal samples up to the pre-defined batch size are selected for a query. Besides that, two approximated approaches are proposed to solve the sparse problem via greedy search and quadratic programming (QP), respectively. After selection, the SVM classifiers are re-trained with new labeled data and the performance is tested on the testing dataset. We conduct several experiments on three image datasets for image classification task. The experimental results show the proposed method outperforms other four different methods and achieves promising performance.