Just Noticeable Difference Values Research Articles

Design and Psychophysical Evaluation of a Novel Wearable Upper-Arm Tactile Display Device.

A novel wearable upper arm tactile display device, which can simultaneously provide three types of tactile stimuli (i.e., squeezing, stretching, and vibration) is presented. The squeezing and stretching stimulation of the skin is generated by two motors simultaneously driving the nylon belt in the opposite and the same direction, respectively. In addition, four evenly spaced vibration motors are fixed around the user's arm by an elastic nylon band. There is also a unique structural design for assembling the control module and actuator, powered by two lithium batteries, making it portable and wearable. Psychophysical experiments are conducted to investigate the effect of interference on the perception of squeezing and stretching stimulation by this device. Results show that (1) different tactile stimuli actually interfere with the user's perception compared to the case where only one stimulus is applied to the user; (2) the squeezing has a considerable impact on the stretch just noticeable difference (JND) values when both stimuli are exerted on the user, and when the squeezing is strong, while the impact of stretch on the squeezing JND values is negligible.

Just noticeable difference of loudness through building wall of masonry and drywall

In this study, the just noticeable difference (JND) of loudness with respect to sound transmission through walls corresponding to a weighted noise reduction index Rw of 43 dB is investigated. Two types of walls typically used in Argentina are evaluated. One is masonry composed of solid autoclaved aerated concrete bricks with a thickness of 25 cm, and the other is a drywall composed of two 12.5 mm plasterboards with a 25 cm air chamber in between them, filled with glass wool. The JND is determined using the method of constant stimuli for both ascending and descending variations. The source signals used are two types of vacuum-cleaner sounds and two music signals. Online tests are conducted owing to restrictions imposed by the COVID-19, and the presentation level of the stimuli is adjusted based on the sensation level of each individual. Responses from 148 different listeners show no significant difference in the subjective responses among the source signals. Regarding the masonry, the JND values are 3.02 and 1.62 cm for the ascending and descending variations, respectively (corresponding Rw difference of 2 and 1 dB), whereas they are 4.55 and 1.26 cm for the drywall, respectively (1 dB for both).

Perceptual Image Compression with Block-Level Just Noticeable Difference Prediction

A block-level perceptual image compression framework is proposed in this work, including a block-level just noticeable difference (JND) prediction model and a preprocessing scheme. Specifically speaking, block-level JND values are first deduced by utilizing the OTSU method based on the variation of block-level structural similarity values between two adjacent picture-level JND values in the MCL-JCI dataset. After the JND value for each image block is generated, a convolutional neural network–based prediction model is designed to forecast block-level JND values for a given target image. Then, a preprocessing scheme is devised to modify the discrete cosine transform coefficients during JPEG compression on the basis of the distribution of block-level JND values of the target test image. Finally, the test image is compressed by the max JND value across all of its image blocks in the light of the initial quality factor setting. The experimental results demonstrate that the proposed block-level perceptual image compression method is able to achieve 16.75% bit saving as compared to the state-of-the-art method with similar subjective quality. The project page can be found at https://mic.tongji.edu.cn/43/3f/c9778a148287/page.htm.

The just noticeable difference of early decay time (EDT)

The purpose of this study was to determine the just noticeable difference (JND) of early decay time (EDT) for broadband conditions and individual frequency bands. JNDs of room acoustic parameters are important for assessing if changes in geometry or materials in a room will have a discernible impact in the listener’s perception. The EDT JND has been reported to be 5% in the ISO 3382-1:2009 standard, which is the same value usually quoted for the JND of classical reverberation time (T30). No formal studies have been undertaken so far to validate if this JND value is appropriate for EDT, or to determine its dependency on frequency. Measured spatial room impulse responses (SRIRs) from three concert halls were used as base cases, which were then modified with a computer algorithm to attain specified values of EDT. All stimuli were reproduced in an anechoic chamber with an array of 30 loudspeakers using third-order ambisonics. The method of constant stimuli, in which a psychometric function is fitted to the measured data, was used to determine the JND. Thirty subjects with musical backgrounds participated in the study. The obtained broadband JND was 26%, while individual frequency band JNDs were higher.

Objective and subjective acoustics measurement of audience seating areas in a medium size auditorium

Several consideration of the acoustical quality of a multi-purpose hall should be taken into account and therefore, the audiences must be fully aware of the acoustical consequences in deciding their seating positions. In this paper, objective measurement and subjective assessment in the medium capacity, Driyarkara multi-purpose hall, has been investigated. Room impulse responses (RIR) were taken for several monaural parameter distribution including G, EDT, and C80 at 16 positions referring to ISO 3382-1:2009. The signals were convolved with anechoic sound of musical instruments to reproduce acoustic stimuli for subjects to listen. Stimuli data is reduced to 8 measurement’ positions, which indicated the noticeable difference. These were generated to reproduce binaural sounds for further subjective assessment. The distribution of binaural parameters were observed in 250 Hz, 1 kHz, and 4 kHz to obtain JND (Just Noticeable Difference). The result reveal that this hall contains 3 different acoustic zones with the range of acoustics parameters of G: -8.91 dB – 1.07 dB, EDT: 1.12 s – 1.52 s, C50: -0.11 dB – 5.05 dB, C80: -3.70 dB – 2.03 dB, D50: 31.00 – 60.71, Ts: 62.75 ms – 106.50 ms, IACC: 0.43 s – 0.82 s, Treble Ratio: 0.67 – 0.96 and Bass Ratio: 0.81 – 2.80. Furthermore, this result correlates with the JND value for G: 2.26 dB, Ts: 17.5 ms, Treble ratio: 0.2, Bass Ratio: 0.03 and D50: 16.67.

Perceptual evaluation of measures of spectral variance

In many applications, it is desirable to achieve a signal that is as close as possible to ideal white noise. One example is in the design of an artificial reverberator, whereby there is a need for its lossless prototype output from an impulse input to be perceptually white as much as possible. The Ljung-Box test, the Drouiche test, and the Wiener Entropy-also called the Spectral Flatness Measure-are three well-known methods for quantifying the similarity of a given signal to ideal white noise. In this paper, listening tests are conducted to measure the Just Noticeable Difference (JND) on the perception of white noise, which is the JND between ideal Gaussian white noise and noise with a specified deviation from the flat spectrum. This paper reports the JND values using one of these measures of whiteness, which is the Ljung-Box test. This paper finds considerable disagreement between the Ljung-Box test and the other two methods and shows that none of the methods is a significantly better predictor of listeners' perception of whiteness. This suggests a need for a whiteness test that is more closely correlated to human perception.

Sparse recovery based reversible data hiding method using the human visual system

In this paper, a novel sparse recovery based reversible data hiding (RDH) method using the human visual system (HVS) is presented. To improve the low accuracy of existing predictors, a sparse recovery based predictor is proposed. In the processes of sparse recovery, the most relevant neighbors can be adaptively chosen by using sparse representation to predict the current pixel accurately, and thus the concentrated prediction error histogram (PEH) is built to obtain good embedding performance. Moreover, to overcome the conflict between the embedding order of the traditional RDH method and the evaluation of HVS, a new embedding strategy based on just noticeable difference (JND) is designed. In this strategy, pixels are classified into sensitive and in-sensitive clusters according to JND values, and two corresponding PEHs are built. Accordingly, different inner regions of two PEHs are adjusted to meet the required embedding capacity, and the prediction error expansion (PEE) technique is utilized to embed data. Experimental results prove that the proposed method outperforms the state-of-the-art RDH methods, including JND related methods.

Method for Simulating Dose Reduction in Digital Breast Tomosynthesis

This paper proposes a new method of simulating dose reduction in digital breast tomosynthesis, starting from a clinical image acquired with a standard radiation dose. It considers both signal-dependent quantum and signal-independent electronic noise. Furthermore, the method accounts for pixel crosstalk, which causes the noise to be frequency-dependent, thus increasing the simulation accuracy. For an objective assessment, simulated and real images were compared in terms of noise standard deviation, signal-to-noise ratio (SNR) and normalized noise power spectrum (NNPS). A two-alternative forced-choice (2-AFC) study investigated the similarity between the noise strength of low-dose simulated and real images. Six experienced medical physics specialists participated on the study, with a total of 2 160 readings. Objective assessment showed no relevant trends with the simulated noise. The relative error in the standard deviation of the simulated noise was less than 2% for every projection angle. The relative error of the SNR was less than 1.5%, and the NNPS of the simulated images had errors less than 2.5%. The 2-AFC human observer experiment yielded no statistically significant difference ( =0.84) in the perceived noise strength between simulated and real images. Furthermore, the observer study also allowed the estimation of a dose difference at which the observer perceived a just-noticeable difference (JND) in noise levels. The estimated JND value indicated that a change of 17% in the current-time product was sufficient to cause a noticeable difference in noise levels. The observed high accuracy, along with the flexible calibration, make this method an attractive tool for clinical image-based simulations of dose reduction.

A DCT-Based Total JND Profile for Spatiotemporal and Foveated Masking Effects

In image and video processing fields, Discrete Cosine Transform (DCT)-based just-noticeable difference (JND) profiles have effectively been utilized to remove perceptual redundancies in pictures for compression. In this paper, we solve two problems that are often intrinsic to the conventional DCT-based JND profiles: 1) no foveated masking (FM) JND model has been incorporated in modeling the DCT-based JND profiles and 2) the conventional temporal masking (TM) JND models assume that all moving objects in frames can be well tracked by the eyes and that they are projected on the fovea regions of the eyes, which is not a realistic assumption and may result in poor estimation of JND values for untracked moving objects (or image regions). To solve these two problems, we first propose a generalized JND model for joint effects between TM and FM effects. With this model, called the temporal–foveated masking (TFM) JND model, JND thresholds for any tracked/untracked and moving/still image regions can be elaborately estimated. Finally, the TFM-JND model is incorporated into a total DCT-based JND profile with a spatial contrast sensitivity function, luminance masking, and contrast masking JND models. In addition, we propose a JND adjustment method for our total JND profile to avoid overestimation of JND values for image blocks of fixed sizes with various image characteristics. To validate the effectiveness of the total JND profile, an experiment involving a subjective distortion-visibility assessment has been conducted. The experiment results show that the proposed total DCT-based JND profile yields significant performance improvement with much higher capability of distortion concealment (average 5.6-dB lower PSNR) compared with state-of-the-art JND profiles. The MATLAB source code of the proposed total DCT-based JND profile is publicly available online at https://sites.google.com/site/sunghobaecv/jnd .

Milk fat threshold determination and the effect of milk fat content on consumer preference for fluid milk

Milk consumption in the United States has been in decline since the 1960s. Milk fat plays a critical role in sensory properties of fluid milk. The first objective of this study was to determine the change in percent milk fat needed to produce a detectable or just noticeable difference (JND) to consumers in skim, 1%, 2%, and whole milks. The second objective was to evaluate how milk fat affected consumer preferences for fluid milk. Threshold tests were conducted to determine the JND for each reference milk (skim, 1%, 2%, and whole milk), with a minimum of 60 consumers for each JND. The JND was determined for milks by visual appearance without tasting and tasting without visual cues. Serving temperature effect (4, 8, or 15°C) on tasting JND values were also investigated. The established JND values were then used to conduct ascending forced-choice preference tests with milks. Consumers were assigned to 3 groups based on self-reported milk consumption: skim milk drinkers (n = 59), low-fat milk drinkers (consumed 1% or 2% milk, n = 64), and whole milk drinkers (n = 49). Follow-up interviews were conducted where consumers were asked to taste and explain their preference between milks that showed the most polarization within each consumer segment. Descriptive sensory analysis was performed on the milks used in the follow-up interviews to quantify sensory differences. Visual-only JND were lower than tasting-only JND values. Preference testing revealed 3 distinct preference curves among the consumer segments. Skim milk drinkers preferred skim milk and up to 2% milk fat, but disliked milk higher in fat due to it being "too thick," "too heavy," "flavor and texture like cream," "too fatty," and "looks like half and half." Low-fat milk drinkers preferred 2% milk up to 3.25% (whole milk), but then disliked higher milk fat content. Whole milk drinkers preferred whichever milk was higher in milk fat regardless of how high the fat content was, distinct from skim and low-fat milk drinkers. The findings of this study provide insights on sensory characteristics of milk fat in fluid milk and consumer sensory perception of these properties. These results also provide insights on how the industry might adjust milk fat references for adjusting milk sensory properties to increase milk preference and remain within the standards of identity of milk.

The effect of boundary shape to acoustic parameters

To design a room in term of acoustic, many variables need to be considered such as volume, acoustic characteristics & surface area of material and also boundary shape. Modifying each variable possibly change the sound field character. To find impact of boundary shape, every needed properties is simulated through acoustic prediction software. The simulation is using three models with different geometry (asymmetry and symmetry) to produce certain objective parameters. By applying just noticeable difference (JND), the effect is considered known. Furthermore, individual perception is needed to gain subjective parameter. The test is using recorded speech that is convoluted with room impulse of each model. The result indicates that 84% of participants could not recognize the speech which is emit from different geometry properties. In contrast, JND value of T30 is exceed 5%. But for D50, every model has JND below 5%.

JND-based spatial parameter quantization of multichannel audio signals

In multichannel spatial audio coding (SAC), the accurate representations of virtual sounds and the efficient compressions of spatial parameters are the key to perfect reproduction of spatial sound effects in 3D space. Just noticeable difference (JND) characteristics of human auditory system can be used to efficiently remove spatial perceptual redundancy in the quantization of spatial parameters. However, the quantization step sizes of spatial parameters in current SAC methods are not well correlated with the JND characteristics. It results in either spatial perceptual distortion or inefficient compression. A JND-based spatial parameter quantization (JSPQ) method is proposed in this paper. The quantization step sizes of spatial parameters are assigned according to JND values of azimuths in a full circle. The quantization codebook size of JSPQ was 56.7 % lower than one of the quantization codebooks of MPEG surround. Average bit rate reduction on spatial parameters for standard 5.1-channel signals reached up to approximately 13 % compared with MPEG surround, while preserving comparable subjective spatial quality.

HEVC-Based Perceptually Adaptive Video Coding Using a DCT-Based Local Distortion Detection Probability Model.

Discrete Cosine Transform (DCT)-based just noticeable difference (JND) profiles have widely been applied into human perception-based video coding in order to reduce perceptual redundancy, which is one of the main goals of perceptual video coding (PVC). However, there are two problems for this approach: 1) the JND value of each transform coefficient is estimated for a fixed-sized DCT kernel (e.g., 8 × 8), but flexible coding structures with variable-sized transform units have been utilized in standard video coding frameworks, such high efficiency video coding (HEVC) and 2) the DCT transform coefficients are suppressed by the amounts of JND values for the removal of perceptual redundancy, but the DCT transform coefficients of residues are not sufficiently suppressed due to many small transform coefficient values in mid- and high-frequency regions below the JND values. In order to solve these problems, we propose a more generalized visibility model in the DCT domain, called the DCT-based local distortion detection probability (LDDP) model that can estimate a degree of distortion visibility for any distribution of the transform coefficients of any sized DCT kernel for residues. Furthermore, we propose an HEVC-compliant LDDP-based PVC scheme where transform coefficients are sufficiently suppressed based on the LDDP model. The proposed PVC scheme is implemented in the HEVC Test Model (HM 11.0) reference software to show the effectiveness of the LDDP-based PVC scheme. Objective and subjective tests for encoded test sequences are performed. The experimental results show that the LDDP-based PVC scheme achieves a significant performance improvement of bitrate reduction at the similar visual quality levels compared with the original HM 11.0.

Non-parametric bootstrapping method for measuring the temporal discrimination threshold for movement disorders

Objective. Recent studies have proposed that the temporal discrimination threshold (TDT), the shortest detectable time period between two stimuli, is a possible endophenotype for adult onset idiopathic isolated focal dystonia (AOIFD). Patients with AOIFD, the third most common movement disorder, and their first-degree relatives have been shown to have abnormal visual and tactile TDTs. For this reason it is important to fully characterize each participant’s data. To date the TDT has only been reported as a single value. Approach. Here, we fit individual participant data with a cumulative Gaussian to extract the mean and standard deviation of the distribution. The mean represents the point of subjective equality (PSE), the inter-stimulus interval at which participants are equally likely to respond that two stimuli are one stimulus (synchronous) or two different stimuli (asynchronous). The standard deviation represents the just noticeable difference (JND) which is how sensitive participants are to changes in temporal asynchrony around the PSE. We extended this method by submitting the data to a non-parametric bootstrapped analysis to get 95% confidence intervals on individual participant data. Main results. Both the JND and PSE correlate with the TDT value but are independent of each other. Hence this suggests that they represent different facets of the TDT. Furthermore, we divided groups by age and compared the TDT, PSE, and JND values. The analysis revealed a statistical difference for the PSE which was only trending for the TDT. Significance. The analysis method will enable deeper analysis of the TDT to leverage subtle differences within and between control and patient groups, not apparent in the standard TDT measure.

Analysis of the JND-Suppression Effect in Quantization Perspective for HEVC-based Perceptual Video Coding

Transform-domain JND (Just Noticeable Difference)-based for PVC (Perceptual Video Coding) is often performed in quantization processes to effectively remove perceptual redundancy. This study examined the JND-suppression effects on quantized coefficients of transform in HEVC (High Efficiency Video Coding). To reveal the JND-suppression effect in quantization, the properties of the floor functions were used for modeling the quantized coefficients, and a JNDadjustment process in an HEVC-compliant PVC scheme was used to tune the JND values by analyzing the JND suppression effect. In the experimental results, the bitrate reduction decreases slightly, but the PSNR and perceptual quality are improved significantly when the proposed JND adjustment process is applied.

Just noticeable difference in perception of physical exertion during cycle exercise in young adult men and women

The purpose of this investigation was to describe the just noticeable difference (JND) in perceived exertion during cycle exercise. Males (n = 20) and females (n = 26) (21.4 ± 3.1 year) performed load-incremented cycle exercise to peak intensity. At the end of each minute, subjects rated their overall-body perceived exertion using the OMNI (0-10) rating of perceived exertion (RPE) scale. Individual regression derived the power output (PO) corresponding to RPE 5. This PO served as the standard stimulus (SS). On a separate occasion, four 5-min cycling bouts were performed with 5 min rest between bouts. During bouts 1 and 3 subjects cycled at the SS. During bouts 2 and 4 subjects adjusted the resistance to achieve a level of exertion just noticeably above/below the SS. The difference in final 30-s oxygen consumption (VO2) and PO between each JND bout and the previous SS were the above (JND-A) and below (JND-B) perceived exertion JNDs. JND-A and JND-B were compared between genders and between subjects exhibiting lower versus higher ventilatory threshold (VT) and VO(2PEAK) within genders for VO2 (l · min(-1), %VO(2PEAK)) and PO (W, %SS). JND-B was significantly (P < 0.05) greater than JND-A for VO2 and PO, when expressed in absolute (l · min(-1), W) and relative units (%VO(2PEAK), %SS). Males exhibited greater JND values than females in absolute, but not relative, units. Subjects with lower and higher VT and VO(2)PEAK exhibited similar JND values. The JND can serve as an effective tool to measure perceptual acuity and to determine individual ability to self-regulate prescribed exercise intensities.

A real-time video watermarking algorithm for streaming media

Video watermarking technology is a useful method to track piracy for digital videos. However, most watermarking algorithms cannot be directly applied to H.264 streaming media because of their poor real-time performance, which may cause users to watch the video unsmoothly. In order to solve this problem, a real-time H.264 video watermarking algorithm (RVWA) is proposed. First, the video is segmented before embedding watermark, which makes watermark embedding and video transmission be carried on at the same time. Then, the concept of watermark embedding margin of frame (WEMF) is proposed to shorten the watermark embedding time. By calculating the value of WEMF for each B frame, which is related to its frame bits, suitable B-frames can be selected to embed watermark without decoding it completely. Finally, with the characteristic of variable macro block size in H.264 coding standard, a simplified formula is presented to quickly calculate the just noticeable difference (JND) value of the macro block, and then the macro blocks with larger JND value are selected to embed watermark into their motion vectors, which also provides good invisibility. Experimental results show that the proposed algorithm can find suitable embedding positions and embed watermark quickly with good invisibility and less increase of the bit-rate of video. RVWA algorithm can be used in stream media environment in real time.

Multiple Causal Window Based Reversible Data Embedding

Reversible data embedding is a technique that embeds data into an image in a reversible manner. An important aspect of reversible data embedding is to find embedding area in the image and to embed the data into it. In the conventional reversible techniques, the visual quality is not taken into account which resulted in a poor quality of the embedded images. Hence the histogram modification based reversible data hiding technique using multiple causal windows is proposed which predicts the embedding level with the help of the pixel value, edge value, Just Noticeable Difference(JND) value. Using this data embedding level the data is embedded into the pixels. The pixel level adjustment considering the Human Visual System (HVS) characteristics is also made to reduce the distortion caused by data embedding. This significantly improves the data embedding capacity along with greater visual quality. The proposed method includes three phases: (i).Construction of casual window and calculation of edge and JND values in which the casual window determines the pixel values, the edge and the JND values are calculated (ii).Data embedding which is the process of embedding the data into the original image (iii). Data extractor and image recovery where the original image is recovered and the embedded bits are obtained. The experimental results and performance comparison with other reversible data hiding algorithms are presented to demonstrate the validity of the proposed algorithm. The experimental results show that the Performance of the proposed system on an average shows an accuracy of 95%.

Determination of Absolute Threshold and Just Noticeable Difference in the Sensory Perception of Pungency

Absolute threshold and just noticeable difference (JND) were determined for the perception of pungency using chili pepper in aqueous solutions. Absolute threshold and JND were determined using 2 alternative forced-choice sensory tests tests. High-performance liquid chromatography technique was used to determine capsaicinoids concentration in samples used for sensory analysis. Sensory absolute threshold was 0.050 mg capsaicinoids/kg sample. Five JND values were determined using 5 reference solutions with different capsaicinoids concentration. JND values changed proportionally as capsaicinoids concentration of the reference sample solutions changed. Weber fraction remained stable for the first 4 reference capsaicinoid solutions (0.05, 0.11, 0.13, and 0.17 mg/kg) but changed when the most concentrated reference capsaicinoids solution was used (0.23 mg/kg). Quantification limit for instrumental analysis was 1.512 mg/kg capsaicinoids. Sensory methods employed in this study proved to be more sensitive than instrumental methods. Practical Application: A better understanding of the process involved in the sensory perception of pungency is currently required because "hot" foods are becoming more popular in western cuisine. Absolute thresholds and differential thresholds are useful tools in the formulation and development of new food products. These parameters may help in defining how much chili pepper is required in a formulated product to ensure a perceptible level of pungency, as well as in deciding how much more chili pepper is required in a product to produce a perceptible increase in its pungency.

Just Noticeable Difference in Perception of Physical Exertion: Relation with Ventilatory Breakpoint and Vo2peak

The just noticeable difference (JND) in the sensory response to conjunctively presented stimuli is a widely applied psychophysical measurement. However, little is known about the JND in the perception of physical exertion. PURPOSE: To describe the JND in the perception of physical exertion during submaximal cycle ergometer exercise. METHODS: Recreationally active males (n=20) and females (n=26) (age 21.4±3.1 yr) performed cycle exercise terminating at peak intensity. Rating of perceived exertion [RPE OMNI Cycle (0-10) Scale] was estimated for the overall-body at the end of each min. Individual linear regression was used to derive the power output (PO) and oxygen consumption (VO2) corresponding to an RPE of 5. This PO served as the standard. On a separate occasion four 5-min cycle exercise bouts were performed with 5 min rest between bouts. During bouts 1 and 3 subjects cycled at the standard. During bouts 2 and 4 subjects adjusted the resistance to achieve a level of exertion above or below the standard. The difference in final 30-s VO2 between each JND bout and the previous standard was taken as the physiologic analog of the above (ΔJND-A) and below (ΔJND-B) perceived exertion JND, expressed in absolute (l·min-1) and relative (%VO2PEAK) values. ΔJND-A and ΔJND-B were compared between genders, as well as between subjects exhibiting lower and higher ventilatory breakpoints (Vpt) and VO2PEAK within genders. RESULTS: For the total sample, ΔJND-B was significantly greater (p<0.05) than ΔJND-A when expressed in absolute (0.24±0.20 l·min-1 vs. 0.17±0.13 l·min-1, respectively) and relative values (8.33±6.01 %VO2PEAK vs. 5.90±4.09 %VO2PEAK, respectively). For both males and females, there were no differences in JND values between subjects with lower and higher Vpt or VO2PEAK. CONCLUSION: It is proposed that the VO2 corresponding to a JND above and below a standard stimulus can serve as an effective tool to measure a subject's perceptual acuity. This can be used to determine an individual's ability to accurately self-regulate prescribed exercise intensities. (This is a secondary analysis of a previously reported data set which has been expanded.)