A cortex-like model for animal recognition based on texture using feature-selective hashing

Seifzadeh, Sahar; Faez, Karim

doi:10.1109/iraniancis.2014.6802584

Cited by 2 publications

(3 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their framework was based on Mutch's improved HMAX model and they use the feature-selective hashing scheme with the Nearest Neighborhood (NN) as a classifier. In addition, in our previous work [6], a biologically-inspired model with feature selective hashing was proposed to recognize animals which was applied on KTH database containing 1239 images in 13 classes with photos taken from the animals' wildlife.…”

Section: The Hmax Modelmentioning

confidence: 99%

A Computational Visual Neuroscience Model for Object Recognition

Seifzadeh

Rezaei²,

Farahbakhsh³

2017

Journal of Advanced Medical Sciences and Applied Technologies

Self Cite

View full text Add to dashboard Cite

In this study with the inspirations from both neuroscience and computer science, a combinatorial framework for object recognition was proposed having benefited from the advantages of both biologically-inspired HMAX_S architecture model for feature extraction and Extreme Learning Machine (ELM) as a classifier. HMAX model is a feed-forward hierarchical structure resembling the ventral pathway in the visual cortex of the brain and ELM is a powerful neural network, which randomly chooses hidden nodes and specifies analytically the single-hidden layer. ELM theories conjecture that this randomness may be true for biological learning in animal brains. It should be noted that the principle reason of using ELM is mainly as a result of its biological structure in order to imitate the biological object recognition system of mammalians and partly for its incredible speed which drastically lessens the runtime. Classification results are reported in Caltech101 dataset, at the focal point with its combinatorial framework serving considerable improvements over latest studies in both classification rate (96.39%) and the low runtime (0.417s). A B S T R A C T

show abstract

Section: The Hmax Modelmentioning

confidence: 99%

A Computational Visual Neuroscience Model for Object Recognition

Seifzadeh

Rezaei²,

Farahbakhsh³

2017

Journal of Advanced Medical Sciences and Applied Technologies

Self Cite

View full text Add to dashboard Cite

show abstract

“…On this basis, the collection of animal images and feature extraction are also indispensable. In particular, it is necessary to collect a number of quality and distinctive animal images as a data set for a specific animal . The collection of a large number of data sets not only depends on human resources but also requires the support of relevant sensors, cameras, and other hardware in the field, and can automatically detect them in the field.…”

Section: Discussionmentioning

confidence: 99%

“…This will not only increase the efficiency of the staff, but reduce the situation in which the criminals do not receive corresponding punishment because the concerning members lacks experience. Furthermore, it can track, identify and classify for animals . Image semantic segmentation based on animal currently has little research studies; from the perspective purpose of protection of rare animals, these research studies are urgently needed, and for protecting those endangered or rare animals, it can help save a lot of time cost, and greatly reduces the identification difficulty about corresponding workers.…”

Section: Introductionmentioning

confidence: 99%

Research of animals image semantic segmentation based on deep learning

Liu

et al. 2018

Concurrency and Computation

View full text Add to dashboard Cite

It is imperative for us to develop the technology of image semantic segmentation with the increasing demand in the image processing. Nowadays, the development of deep learning is of great significance to the improvement of image segmentation. Furthermore, the paper discussed the relationship between image semantic segmentation and animal image research based on the actual situation, and found that animal image processing technology plays a more important role in the field of protecting precious animals. The end-to-end network training of this paper is consisted of Fully Convolutional Network (FCN) for the front end and Conditional Random Fields as Recurrent Neural Networks (CRF-RNN) for the back end via comparing a variety of research methods. The experiments achieved desired outcome for the semantic segmentation of animal images by utilizing Caffe deep learning framework and explained the implementation details from the aspects of training and testing. KEYWORDS Caffe, CRF-RNN, FCN, image semantic segmentation INTRODUCTIONTraditionally, the segmentation of the image is only directly on the surface layer, such as the appearance features based on texture, color, and shape, and they are separated after aggregating the similar features. Among them, the segmentation method based on regions, because the selection of seed points will lead to low efficiency of the algorithm, the segmented results have a certain flaw and can not reach the expected effect. However, the histogram-based threshold segmentation method ignores certain image space information, and it is very likely to segment a discontinuous and meaningless regions, making the segmentation result less satisfying. Therefore, with the demand for image segmentation technology, the semantic segmentation is indispensable.In semantic segmentation, the method of deep learning can be well escaped from the traditional processing methods, and the limitation of using only the underlying features is eliminated, making process in the level of understanding for image. The deep learning framework of Caffe 1 can handle semantic segmentation well. As the number of treasured animals continues to decline, the application of science and technology to protect them is imminent. In particular, it is necessary to collect a set of quality and distinctive animal image data for a specific animal, and the collection of a large number of data sets depends not only on human resources but also requires the support of the relevant sensors, cameras, and other hardware in the wild places it can automatically detect in the field or other places. One solution is to establish a corresponding index for the existing animal picture database. For the protected animal, it can use image processing technology to quickly identify for these animals, and then semantic segmentation is exploited to make a good recognition of animal images. This will not only increase the efficiency of the staff, but reduce the situation in which Concurrency Computat Pract Exper. 2020;32:e4892. wileyonlinelibrary.com/journal/cp...

show abstract

A cortex-like model for animal recognition based on texture using feature-selective hashing

Cited by 2 publications

References 8 publications

A Computational Visual Neuroscience Model for Object Recognition

A Computational Visual Neuroscience Model for Object Recognition

Research of animals image semantic segmentation based on deep learning

Contact Info

Product

Resources

About