Humans vibrotactile sensitivity varies with different vibratory stimuli. The frequency response reaches maximum sensitivity in the region of 200-300 Hz. This sensitivity-frequency characteristics has been extensively studied in a series of psychophysical experiments on human glabrous skin. To this end, we design a perceptual codec for vibrotactile signals that models the tactile human sensitivity characteristics which employed to remove perceptually irrelevant infromation in vibrotactile signals. PVC-SLP is currentley under investigation in the IEEE P1918.1.1 Haptic Codecs Task Group. The codec is also based on sparse linear prediction (SLP) coding. Detailed description is given in the following paper:
Finding ways to automatically and accurately predict the tactile content quality as perceived by end-users can largely impact advances in Tactile Internet and haptic telecommunication systems. We focus our attention here to vibrotactile signals which pertain to the roughness submodality of tactile perception. VibroTactile Quality Assessment (VTQA); unlike speech, image, video; is hampered by various hurdles including the plethora of vibrotactile acquisition/display systems that are used in recording and delivering the tactile test materials, as well as aspects related to limited humans sensitivity to high-frequency vibrations of tactile stimuli. That said, we proposed an objective vibrotactile quality assessment measure called Spectral Temporal SIMilarity (ST-SIM) and a novel subjective test procedure. Together serve as a holistic framework for benchmarking Tactile Codecs (TC) in terms of efficiency and transparency. Subjective rate database is made publicaly availabe.
Teleoperation systems use haptic interface, like the Force Dimension Omega.6 in the figure, to transmit force feedback and velocity (haptic data) generated by the physical interaction with objects in remote real or virtual environments. This technology enables a wide variety of applications including remote operations in dangerous environments, long distance medical diagnostics and surgeries, and interactive games. However, haptic data demands real-time processing with strict delay constraints and high update rates up to the sampling rate (1kHz or more). Hence limit the use of any signal processing and compression algorithms that introduce algorithmic delay. In this context, an important question arises : How can one judge and compare the quality of compressed haptic signals for haptic communication?
To this end, a virtual environment application was built and implemented using C++ and the CHAI 3D software library for computer haptics as shown in the Figure. The resulting force, during the interaction with virtual objects, is computed by modelling a virtual spring between the position of the haptic device end-effector and a virtual proxy using the finger-proxy model in CHAI 3D. The force-feedback test sequences were distorted using the perceptual deadband compression technique. The proposed HSSIM, that contains a ”perceptual model” block and a method to quantify the distortion between reference and test force feedback represented by the SSIM, is used to assess force signal quality. More details can be found in:
Image Sharpness Assessment Based on Local Phase Coherence
Sharpness is an important determinant in visual assessment of image quality. The human visual system is able to effortlessly detect blur and evaluate sharpness of visual images, but it still yet a challenging task for computers. Existing blur/sharpness evaluation algorithms are mostly based on edge width, local gradient, or energy reduction of global/local high frequency content. In this work, a no-reference sharpness assessment algorithm is developed based on the identification of sharpness from a different perspective as strong local phase coherence (LPC) near distinctive image features evaluated in the complex wavelet transform domain.
In the Figure above, we show an exmaple of a natural image and its corresponding LPC map. The image has been blurred using Gaussian blur filter with various sigma values. LPC-SI gives a very good estimation of sharpness quality of a given image. Detailed description is given in the following paper: