The contrast sensitivity function for human vision is well understood. It is known that the contrast sensitivity of luminance discrimination shows a band pass profile and that of chromatic discrimination shows a low pass profile. This characteristics of human vision is applied to the NTSC television transmission system. The image on color TV is formed by the basis of additive mixture of three primary colors; red, green and blue. When we estimate the image quality of color TV, therefore, it becomes very important to know the spatial frequency characteristics of each primary color for human vision.
{"title":"Detection of spatial-frequency selected color shifts and the contrast sensitivity functions for CRT primaries","authors":"H. Yaguchi, Hidemi Takahashi, Y. Miyake","doi":"10.1364/av.1989.fb5","DOIUrl":"https://doi.org/10.1364/av.1989.fb5","url":null,"abstract":"The contrast sensitivity function for human vision is well understood. It is known that the contrast sensitivity of luminance discrimination shows a band pass profile and that of chromatic discrimination shows a low pass profile. This characteristics of human vision is applied to the NTSC television transmission system. The image on color TV is formed by the basis of additive mixture of three primary colors; red, green and blue. When we estimate the image quality of color TV, therefore, it becomes very important to know the spatial frequency characteristics of each primary color for human vision.","PeriodicalId":344719,"journal":{"name":"Applied Vision","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133694895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Images filtered with bandpass filters preserve most of their original characteristics. Quantifying them, we recover the original intensity and reflectance representations from the bandpassed data.
用带通滤波器滤波后的图像保留了大部分原始特征。量化它们，我们从带通数据中恢复原始强度和反射率表示。
{"title":"Image Characteristics Recovery from Bandpass Filtering","authors":"R. Alter-Gartenberg, R. Narayanswamy","doi":"10.1364/av.1989.wd3","DOIUrl":"https://doi.org/10.1364/av.1989.wd3","url":null,"abstract":"Images filtered with bandpass filters preserve most of their original characteristics. Quantifying them, we recover the original intensity and reflectance representations from the bandpassed data.","PeriodicalId":344719,"journal":{"name":"Applied Vision","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125422724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A multirate filter bank (MRB) consists of a set of filters that produce several reduced sampling-rate versions of the same input signal (fig. 1). They have been found useful for subband coding [1, 2], image compression [3] and other areas [3, 4]. The possible implementations for MRBs tradeoff computational with connectional complexities and handle processor noise differently. We explore the consequences of the various MRB implementations for retinal processing and show that fully parallel processing or hybrid parallel/hierarchic processing are candidates for the retina.
{"title":"A Multirate Filter Bank Perspective of Retinal Processing","authors":"B. Levitan, G. Buchsbaum","doi":"10.1364/av.1989.wb3","DOIUrl":"https://doi.org/10.1364/av.1989.wb3","url":null,"abstract":"A multirate filter bank (MRB) consists of a set of filters that produce several reduced sampling-rate versions of the same input signal (fig. 1). They have been found useful for subband coding [1, 2], image compression [3] and other areas [3, 4]. The possible implementations for MRBs tradeoff computational with connectional complexities and handle processor noise differently. We explore the consequences of the various MRB implementations for retinal processing and show that fully parallel processing or hybrid parallel/hierarchic processing are candidates for the retina.","PeriodicalId":344719,"journal":{"name":"Applied Vision","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129187632","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
There is a rapidly growing awareness that the quality experience of the human observer is the ultimate criterion for the technical quality of display systems, coding algorithms and image-processing systems.
人们迅速认识到，人类观察者的经验质量是显示系统、编码算法和图像处理系统技术质量的最终标准。
{"title":"Perceptual Image Quality Metrics","authors":"J. Roufs, H. de Ridder, J. Westerink","doi":"10.1364/av.1989.tha2","DOIUrl":"https://doi.org/10.1364/av.1989.tha2","url":null,"abstract":"There is a rapidly growing awareness that the quality experience of the human observer is the ultimate criterion for the technical quality of display systems, coding algorithms and image-processing systems.","PeriodicalId":344719,"journal":{"name":"Applied Vision","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122803178","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In previous experiments we have attempted to measure color differences required to obtain parallel visual search. (Nagy and Sanchez, 1988) The observer's task was to search for a small target disk that differed only in color from white distractor disks also present in the display. When the color difference between the target and distractors was small the time required to find the target depended on the number of distractors present. We refer to this as a serial search. When the color difference was large the search time was independent of the number of distractors present. We refer to this as parallel search. In order to estimate the color difference required to obtain parallel search we measured search time as a function of color difference in displays with a large number of distractors. Search times decreased with increases in color difference until a minimum was reached and then remained constant with further increases in color difference. From these experiments we estimated the smallest color difference at which parallel search could be obtained along several lines in color space. We will refer to these color differences as critical color differences, or CCD's. CCD's varied from approximately 12 to 40 times the size of the MacAdam unit of color difference depending on the direction in color space.
在以前的实验中，我们尝试测量获得平行视觉搜索所需的色差。(Nagy and Sanchez, 1988)观察者的任务是寻找一个小的目标圆盘，它与显示器上的白色干扰圆盘只有颜色不同。当目标和干扰物的色差较小时，发现目标所需的时间取决于存在的干扰物的数量。我们称之为连续搜索。当色差较大时，搜索时间与存在的干扰物数量无关。我们称之为并行搜索。为了估计获得并行搜索所需的色差，我们测量了具有大量干扰物的显示器的搜索时间作为色差的函数。搜索时间随着色差的增加而减少，直到达到最小值，然后随着色差的进一步增加而保持不变。从这些实验中，我们估计了在色彩空间中沿着几条线进行并行搜索时的最小色差。我们将这些色差称为临界色差(CCD)。CCD的大小大约是MacAdam色差单位的12到40倍，这取决于色彩空间的方向。
{"title":"Color and Visual Search in Large and Small Display Fields","authors":"A. Nagy, Robert R. Sanchez, Thomas C. Hughes","doi":"10.1364/av.1989.fb2","DOIUrl":"https://doi.org/10.1364/av.1989.fb2","url":null,"abstract":"In previous experiments we have attempted to measure color differences required to obtain parallel visual search. (Nagy and Sanchez, 1988) The observer's task was to search for a small target disk that differed only in color from white distractor disks also present in the display. When the color difference between the target and distractors was small the time required to find the target depended on the number of distractors present. We refer to this as a serial search. When the color difference was large the search time was independent of the number of distractors present. We refer to this as parallel search. In order to estimate the color difference required to obtain parallel search we measured search time as a function of color difference in displays with a large number of distractors. Search times decreased with increases in color difference until a minimum was reached and then remained constant with further increases in color difference. From these experiments we estimated the smallest color difference at which parallel search could be obtained along several lines in color space. We will refer to these color differences as critical color differences, or CCD's. CCD's varied from approximately 12 to 40 times the size of the MacAdam unit of color difference depending on the direction in color space.","PeriodicalId":344719,"journal":{"name":"Applied Vision","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116845952","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Computer Graphics has always drawn on vision research although often rather indirectly. From the beginning, issues such as flicker fusion and resolution had to be considered in designing systems for computer graphics. However, there was a large body of experience and well-established standards from the cinema and television industries to provide guidelines. Later, when research focused more on the attainment of realism in images computed from numeric models, issues were raised which required a more basic understanding of vision. Currently, while the community, as a whole, is aware of the importance of underlying principles of vision, there is little reference to vision research in journals and conferences on computer graphics. On the other hand, many advances in computer graphics were based on results found in the vision literature. A few examples follow.
{"title":"Vision Research Applied to Computer-Synthesized Imagery","authors":"F. Crow","doi":"10.1364/av.1989.wc2","DOIUrl":"https://doi.org/10.1364/av.1989.wc2","url":null,"abstract":"Computer Graphics has always drawn on vision research although often rather indirectly. From the beginning, issues such as flicker fusion and resolution had to be considered in designing systems for computer graphics. However, there was a large body of experience and well-established standards from the cinema and television industries to provide guidelines. Later, when research focused more on the attainment of realism in images computed from numeric models, issues were raised which required a more basic understanding of vision. Currently, while the community, as a whole, is aware of the importance of underlying principles of vision, there is little reference to vision research in journals and conferences on computer graphics. On the other hand, many advances in computer graphics were based on results found in the vision literature. A few examples follow.","PeriodicalId":344719,"journal":{"name":"Applied Vision","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123639769","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alphanumeric characters often appear distorted when displayed on cathode ray tube devices. These image distortions, referred to as "jaggies", result from undersampling the original high-resolution versions of the characters. To eliminate the aliasing errors that result from undersampling an image, one can filter out frequencies greater than the Nyquist limit - however practical limitation of the display device (gaussian pixel profile and limited intensity range) make the filtering less than ideal. (Kajiya, 1981) Additional image distortions occur because of a mismatch between the filters used to sample a high-resolution image (referred to as convolution kernels) and the pixel point-spread function of the monitor used to display the sampled and filtered image (referred to as the reconstruction kernel).
{"title":"An Image Quality Metric for Digital Letterforms","authors":"J. Farrell, Andrew Fitzhugh","doi":"10.1364/av.1989.thb5","DOIUrl":"https://doi.org/10.1364/av.1989.thb5","url":null,"abstract":"Alphanumeric characters often appear distorted when displayed on cathode ray tube devices. These image distortions, referred to as \"jaggies\", result from undersampling the original high-resolution versions of the characters. To eliminate the aliasing errors that result from undersampling an image, one can filter out frequencies greater than the Nyquist limit - however practical limitation of the display device (gaussian pixel profile and limited intensity range) make the filtering less than ideal. (Kajiya, 1981) Additional image distortions occur because of a mismatch between the filters used to sample a high-resolution image (referred to as convolution kernels) and the pixel point-spread function of the monitor used to display the sampled and filtered image (referred to as the reconstruction kernel).","PeriodicalId":344719,"journal":{"name":"Applied Vision","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121098790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The visual system is trichromatic for small, centrally fixated fields and high (photopic) light intensities. For such conditions, color matching requires three primary lights, because of the three kinds of cone photoreceptors involved. When light intensities are low (scotopic), color matching reduces to brightness matching, and vision becomes monochromatic. At scotopic light intensities, which are not enough to excite cones, vision is mediated by rods; on the other hand, photopic light intensities are high enough to saturate the rods so that only cones contribute to color matches. But what of light intensities (mesopic) that are above the cone threshold but below the rod-saturation intensity? If rods and cones participate together in mesopic vision, shouldn't mesopic color matches require four primary lights instead of three? To address this question, this paper first shows the possibility of a full tetrachromatic matching space by demonstrating that the rod spectral-sensitivity function is linearly independent of a set of color-matching functions for the cones. Secondly, the mathematical conditions for convergence of Trezona's [1] iterative tetrachromatic matching experiment are derived and shown to agree with the actual conditions of the experiment. This step is intended to show that some experimental results are consistent with the formal discussion. Finally, a simple diagram is introduced that represents confusion loci in a reduced "matching space" of rod versus cone stimulation. The topology of this diagram is used to show how mesopic trichromacy is possible.
{"title":"Theoretical Constraints on the Participation of Rods and Cones in Color Matches","authors":"M. H. Brill","doi":"10.1364/av.1989.fb3","DOIUrl":"https://doi.org/10.1364/av.1989.fb3","url":null,"abstract":"The visual system is trichromatic for small, centrally fixated fields and high (photopic) light intensities. For such conditions, color matching requires three primary lights, because of the three kinds of cone photoreceptors involved. When light intensities are low (scotopic), color matching reduces to brightness matching, and vision becomes monochromatic. At scotopic light intensities, which are not enough to excite cones, vision is mediated by rods; on the other hand, photopic light intensities are high enough to saturate the rods so that only cones contribute to color matches. But what of light intensities (mesopic) that are above the cone threshold but below the rod-saturation intensity? If rods and cones participate together in mesopic vision, shouldn't mesopic color matches require four primary lights instead of three? To address this question, this paper first shows the possibility of a full tetrachromatic matching space by demonstrating that the rod spectral-sensitivity function is linearly independent of a set of color-matching functions for the cones. Secondly, the mathematical conditions for convergence of Trezona's [1] iterative tetrachromatic matching experiment are derived and shown to agree with the actual conditions of the experiment. This step is intended to show that some experimental results are consistent with the formal discussion. Finally, a simple diagram is introduced that represents confusion loci in a reduced \"matching space\" of rod versus cone stimulation. The topology of this diagram is used to show how mesopic trichromacy is possible.","PeriodicalId":344719,"journal":{"name":"Applied Vision","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129044858","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Measurements of contrast sensitivity for "jutter" perception are described as a function of sampling frequency, spatial frequency and velocity for luminance and isoluminant chromaticity gratings.
对比敏感度的测量“抖动”感知被描述为一个函数的采样频率，空间频率和速度的亮度和等亮度色度光栅。
{"title":"Perception of \"Jutter\" in Temporally Sampled Images","authors":"W. Glenn, K. Glenn","doi":"10.1364/av.1989.thc1","DOIUrl":"https://doi.org/10.1364/av.1989.thc1","url":null,"abstract":"Measurements of contrast sensitivity for \"jutter\" perception are described as a function of sampling frequency, spatial frequency and velocity for luminance and isoluminant chromaticity gratings.","PeriodicalId":344719,"journal":{"name":"Applied Vision","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129946307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Over about the last two decades the use of digitally generated halftone images has increased significantly. Halftone images which are created digitally from sampled and quantized images have some unique characteristics which affect the perceived image quality. Among these are the number of gray levels which can be reproduced at any spatial frequency, how the screen pattern chosen affects those gray levels, how tone reproduction is controlled and how the structured noise affects appearance. This paper provides an overview of some of the work on these effects.
{"title":"Quality Measures in Digital Halftones","authors":"P. Roetling","doi":"10.1364/av.1989.tha3","DOIUrl":"https://doi.org/10.1364/av.1989.tha3","url":null,"abstract":"Over about the last two decades the use of digitally generated halftone images has increased significantly. Halftone images which are created digitally from sampled and quantized images have some unique characteristics which affect the perceived image quality. Among these are the number of gray levels which can be reproduced at any spatial frequency, how the screen pattern chosen affects those gray levels, how tone reproduction is controlled and how the structured noise affects appearance. This paper provides an overview of some of the work on these effects.","PeriodicalId":344719,"journal":{"name":"Applied Vision","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132078921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：