The key frames extraction technique extracts key postures to describe the original motion sequence, which has been widely used in motion compression, motion retrieval, motion edition and so on. In this paper, we propose a method based on the amplitude of curve to find key frames in a motion captured sequence. First we select a group of joint distance features to represent the motion and adopt the Principal Component Analysis (PCA) method to obtain the one dimension principal component as a features curve which will be used. Then we gain the initial key-frames by extracting the local optimum points in the curve. At last, we get the final key frames by inserting frames based on the amplitude of the curve and merging key frames too close. A number of experimental examples demonstrate that our method is practicable and efficient not only in the visual performance but also in the aspect of the compression ratio and error rate.

No abstract

No abstract

For the problem of Gaussian noise and impulse noise co-exist in acoustic image collected by AUV sonar, based on the improved pulse coupled neural network model, a hybrid filter is proposed using the advantages of median and wiener filtering. It is indicated that captures among neurons act on image filtering to a certain extent through research on the work principle of pulse coupled neural network, and the capability and characteristic of filtering noises are discussed. For filtered sonar image, edge is detected by selecting complex way of generalized morphology opening-closing operation. Then experimental results are given which show that the filter performance of the proposed approach is significantly superior to traditional filter approaches and the proposed morphology edge detection approach can ensure the continuity, integrity and accurate positioning of the edge of underwater sonar image. In a word, it is indicated that the proposed approach can provide effective environment observation information for AUV autonomous navigation.

No abstract