Image Processing Strategies Based on a Visual Saliency Model for Object Recognition Under Simulated Prosthetic Vision

Wang, Jing; Li, Heng; Fu, Weizhen; Yao, Chen; Li, Liming; Lyu, Qing; Han, Tingting; Chai, Xinyu

doi:10.1111/aor.12498

Cited by 25 publications

(27 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Under narrow activation spread conditions, stimulation of the retina takes A number of strategies have been proposed to achieve a desired target pattern of retinal activation. The conventional stimulation strategy sets electrode amplitudes to correspond to the desired target activations at their locations on the retina [12][13][14][15][16][17][18][19]. Under this strategy, the electrode activation pattern is simply a down-sampled (pixelated) version of the desired retinal activation pattern, with the amplitude typically scaled between the threshold and maximum allowed levels.…”

Section: Introductionmentioning

confidence: 99%

Global activity shaping strategies for a retinal implant

et al. 2019

View full text Add to dashboard Cite

Objective: Retinal prostheses provide visual perception via electrical stimulation of the retina using an implanted array of electrodes. The retinal activation resulting from each electrode is not point-like; instead each electrode introduces a spread of retinal activation that may overlap with activations from other electrodes. With most conventional stimulation strategies this overlap leads to image blur. Here we propose a 'shaping' algorithm that uses multiple electrodes to manipulate the current between electrodes in a desired way. Approach: We assume a forward model for the conversion of electrode strengths to retinal activation. Three alternative global shaping algorithms are developed by calculating reverse models under different assumptions: linear inversion using singular value decomposition to 2 produce the pseudoinverse, a linearly constrained quadratic program, and a binary quadratic program to partition the target pattern. The algorithms were assessed using both the mean squared error between the resulting images and desired images, as well as their adherence to the maximum allowed electrode currents. Main Results: Under wide activation spreads the linear inversion algorithm gave improved solutions but faced two limitations: under low-noise conditions the electrode amplitudes exceeded their set limit; the set of solutions did not include the possibility of using negative local currents to induce retinal activation. The linearly constrained quadratic program and binary quadratic program respectively addressed these problems, but required much greater computation time. Significance: This provides a framework for improving the resolution of future retinal implants, especially those with high density electrode arrays.

show abstract

Section: Introductionmentioning

confidence: 99%

Global activity shaping strategies for a retinal implant

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The saliency detection models are based on the visual attention mechanism and are used to extract the salient features to generate the saliency map. Common models such as the Itti and the GBVS are widely used in the field of visual prosthesis [ 12 , 13 ]. However, the saliency map detected by the common model is discrete region [ 15 , 16 ].…”

Section: Image Processing Strategies Based On Salient Object Detecmentioning

confidence: 99%

“…Their results demonstrated that the saliency map can provide clues for searching and performing tasks for users with visual prosthesis. Wang et al [ 12 ] and Li et al [ 13 ] proposed two image processing strategies based on improved Itti and GBVS model to optimize the presentation in simulated prosthetic vision, respectively. Their results demonstrated that the use of the saliency segmentation method and image processing strategies can automatically extract and enhance foreground objects and significantly improve object recognition performance towards recipients with a high-density implant.…”

Section: Introductionmentioning

confidence: 99%

Optimization of Visual Information Presentation for Visual Prosthesis

Guo

Yang

Gao

2018

International Journal of Biomedical Imaging

View full text Add to dashboard Cite

Visual prosthesis applying electrical stimulation to restore visual function for the blind has promising prospects. However, due to the low resolution, limited visual field, and the low dynamic range of the visual perception, huge loss of information occurred when presenting daily scenes. The ability of object recognition in real-life scenarios is severely restricted for prosthetic users. To overcome the limitations, optimizing the visual information in the simulated prosthetic vision has been the focus of research. This paper proposes two image processing strategies based on a salient object detection technique. The two processing strategies enable the prosthetic implants to focus on the object of interest and suppress the background clutter. Psychophysical experiments show that techniques such as foreground zooming with background clutter removal and foreground edge detection with background reduction have positive impacts on the task of object recognition in simulated prosthetic vision. By using edge detection and zooming technique, the two processing strategies significantly improve the recognition accuracy of objects. We can conclude that the visual prosthesis using our proposed strategy can assist the blind to improve their ability to recognize objects. The results will provide effective solutions for the further development of visual prosthesis.

show abstract

“…Han et al [11] utilized feature extraction and image enhancement strategy to improve the accuracy and efficiency of object recognition. Aimed at highlighting the main object of a normal image, two different ways of pixelization [12] proved to be beneficial in daily object recognition tasks.…”

Section: Introductionmentioning

confidence: 99%

“…This study focuses on realizing the goal of obstacle avoidance for the blind based on existing visual protheses. Unlike previous studies [9], [10], [11], [12] which tried to optimize the content of phosphenes in order to improve the performances of visual tasks for the blind, our NeuCube-based obstacle avoidance system directly tells the blind the classification result of obstacle analysis, without any interaction between the prothesis wearers and the systems. As an extension of existing visual prothesis, the proposed obstacle avoidance method can make use of the down-sampled signal from information processing unit of visual prothesis, providing useful guiding information to the blind.…”

Section: Introductionmentioning

confidence: 99%

A spiking neural network model for obstacle avoidance in simulated prosthetic vision

Kasabov

et al. 2017

Information Sciences

View full text Add to dashboard Cite

Limited by visual percepts elicited by existing visual prothesis, it's necessary to enhance its functionality to fulfill some challenging tasks for the blind such as obstacle avoidance. This paper provides a new methodology for obstacle avoidance in simulated prosthetic vision by modelling and classifying spatiotemporal (ST) video data. The proposed methodology is based on a novel spiking neural network architecture, called NeuCube as a general framework for video data modelling in simulated prosthetic vision. As an integrated environment including spiking trains encoding, input variable mapping, unsupervised reservoir training and supervised classifier training, the NeuCube consists of a spiking neural network reservoir (SNNr) and a dynamic evolving spiking neural network classifier (deSNN). First input is captured by visual prothesis, then ST feature extraction is utilized in the low-resolution prosthetic vision generated by the prothesis. Finally such ST features are fed to the NeuCube to output classification result of obstacle analysis. Experiments on collected video data and comparison with other computational intelligence methods indicate promising results. The proposed NeuCube-based obstacle avoidance methodology provides useful guidance to the blind, thus improving the current prothesis and hopefully benefiting the future prothesis wearers.

show abstract

Image Processing Strategies Based on a Visual Saliency Model for Object Recognition Under Simulated Prosthetic Vision

Cited by 25 publications

References 24 publications

Global activity shaping strategies for a retinal implant

Global activity shaping strategies for a retinal implant

Optimization of Visual Information Presentation for Visual Prosthesis

A spiking neural network model for obstacle avoidance in simulated prosthetic vision

Contact Info

Product

Resources

About