It's not easy being green: student recall of plant and animal images

It is well documented that people are less interested in studying plants than animals. We tested whether university students would selectively recall more animal images than plant images even when equally-nameable plant and animal images were presented for equal lengths of time. Animal and plant images were pre-tested and 14 animal-plant pairs were selected, based on student ability to equally name the images. These images were randomly presented to two groups of university students: those currently enrolled in a psychology class and those currently enrolled in a botany class. Student recall of each image was recorded after a distracting task. The results confirmed that the animal images were recalled significantly more than the plant images. There was no apparent effect of attending a botany class on these results. However, gender effects were identified for recall of plant versus animal images in general (women recalled more plants than men) and for four specific plant images (carnation, rose, daisy, and venus fly trap). When teaching biology, teachers should present equal numbers of plant and animal examples and use the most memorable plant images possible to attempt to offset student selective attention to animals.     

Conditional generative adversarial networks to generate pseudo low monoenergetic CT image from a single-tube voltage CT scanner

PurposeTo generate pseudo low monoenergetic CT images of the abdomen from 120-kVp CT images with cGAN.Materials and MethodsWe retrospectively included 48 patients who underwent contrast-enhanced abdominal CT using dual-energy CT. We reconstructed paired data sets of 120 kVp CT images and virtual low monoenergetic (55-keV) CT images. cGAN was prepared to generate pseudo 55-keV CT images from 120-kVp CT images. The pseudo 55 keV CT images in epoch 10, 50, 100, and 500 were compared to the 55 keV images generated using peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM).ResultsThe PSNRs were 28.0, 28.5, 28.6, and 28.8 at epochs 10, 50, 100, and 500, respectively. The SSIM was approximately constant from epochs 50 to 500.ConclusionPseudo low monoenergetic abdominal CT images were generated from 120-kVp CT images using cGAN, and the images had good quality similar to that of monochromatic images obtained with DECT software.     

An automatic method for removing empty camera trap images using ensemble learning

Camera traps often produce massive images, and empty images that do not contain animals are usually overwhelming. Deep learning is a machine‐learning algorithm and widely used to identify empty camera trap images automatically. Existing methods with high accuracy are based on millions of training samples (images) and require a lot of time and personnel costs to label the training samples manually. Reducing the number of training samples can save the cost of manually labeling images. However, the deep learning models based on a small dataset produce a large omission error of animal images that many animal images tend to be identified as empty images, which may lead to loss of the opportunities of discovering and observing species. Therefore, it is still a challenge to build the DCNN model with small errors on a small dataset. Using deep convolutional neural networks and a small‐size dataset, we proposed an ensemble learning approach based on conservative strategies to identify and remove empty images automatically. Furthermore, we proposed three automatic identifying schemes of empty images for users who accept different omission errors of animal images. Our experimental results showed that these three schemes automatically identified and removed 50.78%, 58.48%, and 77.51% of the empty images in the dataset when the omission errors were 0.70%, 1.13%, and 2.54%, respectively. The analysis showed that using our scheme to automatically identify empty images did not omit species information. It only slightly changed the frequency of species occurrence. When only a small dataset was available, our approach provided an alternative to users to automatically identify and remove empty images, which can significantly reduce the time and personnel costs required to manually remove empty images. The cost savings were comparable to the percentage of empty images removed by models.     

Gradient direction dependencies in natural images

We have used information-theoretic measures to compute the amount of dependency which exists between two and three gradient directions at separate locations in an ensemble of natural images. Control experiments were performed on other image classes: phase randomized natural images, whitened natural images and Gaussian noise images. The results show that, for an ensemble of natural images, the amount of 2-point and 3-point gradient direction dependency is equivalent to its ensemble of phase randomized natural images. Therefore, we conclude that the amount of gradient direction dependency in an ensemble of natural images is determined by the ensemble's mean power spectrum rather than the phase spectra of the images. Moreover, this relationship does not extend to individual natural images, the amount of dependency between gradient magnitudes, or gradient directions at high gradient magnitude locations.     

Grayscale representation of infrared microscopy images by extended multiplicative signal correction for registration with histological images

Fourier‐transform infrared (FTIR) microspectroscopy is rounding the corner to become a label‐free routine method for cancer diagnosis. In order to build infrared‐spectral based classifiers, infrared images need to be registered with Hematoxylin and Eosin (H&E) stained histological images. While FTIR images have a deep spectral domain with thousands of channels carrying chemical and scatter information, the H&E images have only three color channels for each pixel and carry mainly morphological information. Therefore, image representations of infrared images are needed that match the morphological information in H&E images. In this paper, we propose a novel approach for representation of FTIR images based on extended multiplicative signal correction highlighting morphological features that showed to correlate well with morphological information in H&E images. Based on the obtained representations, we developed a strategy for global‐to‐local image registration for FTIR images and H&E stained histological images of parallel tissue sections.     

Fractional Vegetation Cover Estimation Based on an Improved Selective Endmember Spectral Mixture Model

Vegetation is an important part of ecosystem and estimation of fractional vegetation cover is of significant meaning to monitoring of vegetation growth in a certain region. With Landsat TM images and HJ-1B images as data source, an improved selective endmember linear spectral mixture model (SELSMM) was put forward in this research to estimate the fractional vegetation cover in Huangfuchuan watershed in China. We compared the result with the vegetation coverage estimated with linear spectral mixture model (LSMM) and conducted accuracy test on the two results with field survey data to study the effectiveness of different models in estimation of vegetation coverage. Results indicated that: (1) the RMSE of the estimation result of SELSMM based on TM images is the lowest, which is 0.044. The RMSEs of the estimation results of LSMM based on TM images, SELSMM based on HJ-1B images and LSMM based on HJ-1B images are respectively 0.052, 0.077 and 0.082, which are all higher than that of SELSMM based on TM images; (2) the R² of SELSMM based on TM images, LSMM based on TM images, SELSMM based on HJ-1B images and LSMM based on HJ-1B images are respectively 0.668, 0.531, 0.342 and 0.336. Among these models, SELSMM based on TM images has the highest estimation accuracy and also the highest correlation with measured vegetation coverage. Of the two methods tested, SELSMM is superior to LSMM in estimation of vegetation coverage and it is also better at unmixing mixed pixels of TM images than pixels of HJ-1B images. So, the SELSMM based on TM images is comparatively accurate and reliable in the research of regional fractional vegetation cover estimation.     

Identification of Cryptosporidium parvum oocysts by an artificial neural network approach

Microscopic detection of Cryptosporidium parvum oocysts is time-consuming, requires trained analysts, and is frequently subject to significant human errors. Artificial neural networks (ANN) were developed to help identify immunofluorescently labeled C. parvum oocysts. A total of 525 digitized images of immunofluorescently labeled oocysts, fluorescent microspheres, and other miscellaneous nonoocyst images were employed in the training of the ANN. The images were cropped to a 36- by 36-pixel image, and the cropped images were placed into two categories, oocyst and nonoocyst images. The images were converted to grayscale and processed into a histogram of gray color pixel intensity. Commercially available software was used to develop and train the ANN. The networks were optimized by varying the number of training images, number of hidden neurons, and a combination of these two parameters. The network performance was then evaluated using a set of 362 unique testing images which the network had never "seen" before. Under optimized conditions, the correct identification of authentic oocyst images ranged from 81 to 97%, and the correct identification of nonoocyst images ranged from 78 to 82%, depending on the type of fluorescent antibody that was employed. The results indicate that the ANN developed were able to generalize the training images and subsequently discern previously unseen oocyst images efficiently and reproducibly. Thus, ANN can be used to reduce human errors associated with the microscopic detection of Cryptosporidium oocysts.     

Stereo super-resolution images detection based on multi-scale feature extraction and hierarchical feature fusion

Recently, with most mobile phones coming with dual cameras, stereo image super-resolution is becoming increasingly popular in phones and other modern acquisition devices, leading stereo super-resolution images spread widely on the Internet. However, current image forensics methods are carried out in monocular images, and high false positive rate appears when detecting stereo super-resolution images by these methods. Therefore, it is important to develop stereo super-resolution image detection method. In this paper, a convolutional neural network with multi-scale feature extraction and hierarchical feature fusion is proposed to detect the stereo super-resolution images. Multi-atrous convolutions are employed to extract multi-scale features and be adapt for varying stereo super-resolution images, and hierarchical feature fusion further improve the performance and robustness of the model. Experimental results demonstrate that the proposed network can detect stereo super-resolution images effectively and achieve strong generalization and robustness. To the best of our knowledge, it is the first attempt to investigate the performance of current forensics methods when tested under stereo super-resolution images, and represent the first study of stereo super-resolution images detection. We believe that it can raise the awareness about the security of stereo super-resolution images.     

Identification of Cryptosporidium parvum Oocysts by an Artificial Neural Network Approach

Microscopic detection of Cryptosporidium parvum oocysts is time-consuming, requires trained analysts, and is frequently subject to significant human errors. Artificial neural networks (ANN) were developed to help identify immunofluorescently labeled C. parvum oocysts. A total of 525 digitized images of immunofluorescently labeled oocysts, fluorescent microspheres, and other miscellaneous nonoocyst images were employed in the training of the ANN. The images were cropped to a 36- by 36-pixel image, and the cropped images were placed into two categories, oocyst and nonoocyst images. The images were converted to grayscale and processed into a histogram of gray color pixel intensity. Commercially available software was used to develop and train the ANN. The networks were optimized by varying the number of training images, number of hidden neurons, and a combination of these two parameters. The network performance was then evaluated using a set of 362 unique testing images which the network had never “seen” before. Under optimized conditions, the correct identification of authentic oocyst images ranged from 81 to 97%, and the correct identification of nonoocyst images ranged from 78 to 82%, depending on the type of fluorescent antibody that was employed. The results indicate that the ANN developed were able to generalize the training images and subsequently discern previously unseen oocyst images efficiently and reproducibly. Thus, ANN can be used to reduce human errors associated with the microscopic detection of Cryptosporidium oocysts.     

TMaCS: A hybrid template matching and classification system for partially-automated particle selection

Selection of particle images from electron micrographs presents a bottleneck in determining the structures of macromolecular assemblies by single particle electron cryomicroscopy (cryo-EM). The problem is particularly important when an experimentalist wants to improve the resolution of a 3D map by increasing by tens or hundreds of thousands of images the size of the dataset used for calculating the map. Although several existing methods for automatic particle image selection work well for large protein complexes that produce high-contrast images, it is well known in the cryo-EM community that small complexes that give low-contrast images are often refractory to existing automated particle image selection schemes. Here we develop a method for partially-automated particle image selection when an initial 3D map of the protein under investigation is already available. Candidate particle images are selected from micrographs by template matching with template images derived from projections of the existing 3D map. The candidate particle images are then used to train a support vector machine, which classifies the candidates as particle images or non-particle images. In a final step in the analysis, the selected particle images are subjected to projection matching against the initial 3D map, with the correlation coefficient between the particle image and the best matching map projection used to assess the reliability of the particle image. We show that this approach is able to rapidly select particle images from micrographs of a rotary ATPase, a type of membrane protein complex involved in many aspects of biology.     

A Technique to Identify Changes in Hemlock Forest Health over Space and Time Using Satellite Image Data

The objective of this study was to develop a technique to classify health of eastern hemlock stands using historical satellite images. While remote sensing and geographic information systems have been used successfully to classify forest health using recent images, applying this process to older images is problematic because contemporaneous field data are not available to measure the accuracy of the classification of historical images. Data ranges were established for each hemlock health class using a contemporary image and field data. These ranges were used to level-slice archived images to create a series of health-class maps that show changes in forest health over time. By applying cross-tabulation procedures to pairs of classified images, it is possible to construct a transition map that indicates how the hemlock health class of each pixel in the images of the study area has changed over time. The resulting maps provide a look back at forest conditions of the past and can be used to identify areas of special interest.     

Image quality characteristics for virtual monoenergetic images using dual-layer spectral detector CT: Comparison with conventional tube-voltage images

PurposeTo investigate the image quality characteristics for virtual monoenergetic images compared with conventional tube-voltage image with dual-layer spectral CT (DLCT).MethodsHelical scans were performed using a first-generation DLCT scanner, two different sizes of acrylic cylindrical phantoms, and a Catphan phantom. Three different iodine concentrations were inserted into the phantom center. The single-tube voltage for obtaining virtual monoenergetic images was set to 120 or 140 kVp. Conventional 120- and 140-kVp images and virtual monoenergetic images (40–200-keV images) were reconstructed from slice thicknesses of 1.0 mm. The CT number and image noise were measured for each iodine concentration and water on the 120-kVp images and virtual monoenergetic images. The noise power spectrum (NPS) was also calculated.ResultsThe iodine CT numbers for the iodinated enhancing materials were similar regardless of phantom size and acquisition method. Compared with the iodine CT numbers of the conventional 120-kVp images, those for the monoenergetic 40-, 50-, and 60-keV images increased by approximately 3.0-, 1.9-, and 1.3-fold, respectively. The image noise values for each virtual monoenergetic image were similar (for example, 24.6 HU at 40 keV and 23.3 HU at 200 keV obtained at 120 kVp and 30-cm phantom size). The NPS curves of the 70-keV and 120-kVp images for a 1.0-mm slice thickness over the entire frequency range were similar.ConclusionVirtual monoenergetic images represent stable image noise over the entire energy spectrum and improved the contrast-to-noise ratio than conventional tube voltage using the dual-layer spectral detector CT.     

Hard X-ray index of refraction tomography of a whole rabbit knee joint: A feasibility study

We report results of the computed tomography reconstruction of the index of refraction in a whole rabbit knee joint examined at the photon energy of 51 keV. Refraction based images make it possible to delineate the bone, cartilage, and soft tissues without adjusting the contrast window width and level. Density variations, which are related to tissue composition and are not visible in absorption X-ray images, are detected in the obtained refraction based images. We discuss why refraction-based images provide better detectability of low contrast features than absorption images.     

基于VTK的医学图像三维可视化系统

医学图像的三维可视化可以通过可视化工具包（VTK）提供的API实现。VTK是医学图像可视化的开法工具包,它把可视化的算法封装起来,利用简单的代码生成所需图形。基于VTK的医学图像三维可视化系统阐述了如何借助VTKAPI读入二维医学图像序列、操作二维图像、重建三维图像以及进行三维图像可视化的全套方案,为临床医生的诊断、治疗提供了有益的途径。     

Animal Detection in Natural Images: Effects of Color and Image Database

The visual system has a remarkable ability to extract categorical information from complex natural scenes. In order to elucidate the role of low-level image features for the recognition of objects in natural scenes, we recorded saccadic eye movements and event-related potentials (ERPs) in two experiments, in which human subjects had to detect animals in previously unseen natural images. We used a new natural image database (ANID) that is free of some of the potential artifacts that have plagued the widely used COREL images. Color and grayscale images picked from the ANID and COREL databases were used. In all experiments, color images induced a greater N1 EEG component at earlier time points than grayscale images. We suggest that this influence of color in animal detection may be masked by later processes when measuring reation times. The ERP results of go/nogo and forced choice tasks were similar to those reported earlier. The non-animal stimuli induced bigger N1 than animal stimuli both in the COREL and ANID databases. This result indicates ultra-fast processing of animal images is possible irrespective of the particular database. With the ANID images, the difference between color and grayscale images is more pronounced than with the COREL images. The earlier use of the COREL images might have led to an underestimation of the contribution of color. Therefore, we conclude that the ANID image database is better suited for the investigation of the processing of natural scenes than other databases commonly used.     

Hippocampal Spatial Position Evaluation on MRI for Research and Clinical Practice

In clinical practice as well as in many volumetric studies we use different reorientations of the brain position towards x and y axis on the magnetic resonance imaging (MRI) scans. In order to find out whether it has an overall effect on the resulting 2D data, manual hippocampal area measurements and rotation variability of the brain (in two reoriented axes) and the skull were performed in 23 Alzheimer''s disease patients and 31 healthy controls. After the MRI scanning, native brain scans (nat) were reoriented into the two different artificial planes (anterior commissure – posterior commissure axis (AC-PC) and hippocampal horizontal long axis (hipp)). Hippocampal area and temporal horn of the lateral ventricle was measured manually using freeware Image J program. We found that 1) hippocampal area of nat images is larger compared to hipp images, area of the nat images is equal to the AC-PC images and area of the hipp images is smaller compared to AC-PC images, 2) hippocampal area together with the area of the temporal horn for nat images is larger compared to hipp images, area of the hipp images is smaller compared to the AC-PC images and area of the nat images is smaller compared to the AC-PC images. The conclusion is that the measured area of the hippocampus in the native MRI is almost the same as the area of MRI reoriented only into the AC-PC axis. Therefore, when performing 2D area studies of the hippocampus or in the clinical practice we recommend usage of not-reoriented MRI images or to reorient them into the AC-PC axis. Surprising finding was that rotation of both AC-PC and hipp line towards x-axis among patients varies up to 35° and the same is true for the skull rotation so that it is not only a matter of the brain position.     

Density-dependent predation by skunks using olfactory search images

The formation of search images can create density-dependent predation. Predators have been shown to form search images when searching for many small prey items in one feeding session. This paper reports experiments that test whether striped skunks can form olfactory search images in other situations: when prey are found over several days, when prey are large, and when prey are found in certain habitats. Striped skunks were raised in captivity, and their reaction distance to food was measured outside in a natural grassy area. In experiment 1 skunks were offered many small food items for several days in a row. From one day to the next, skunks initially detected food from further away, they increased detection distance faster and their maximum detection distance increased – i.e., they formed olfactory search images faster and stronger from one day to the next. In experiment 2 skunks formed search images over several days when finding only one large food item per day. In experiment 3 skunks lost olfactory search images when they entered habitats in which they had previously searched for another type of food. These long-term search images magnify the effects of short-term search images, extend the effects of short- term search image to longer time spans, and affect different species from short-term search images. Received: 26 July 1996 / Accepted: 13 December 1996     

Image quality assessment of an iterative reconstruction algorithm applied to abdominal CT imaging

PurposeTo compare the noise and accuracy on images of the whole porcine liver acquired with iterative reconstruction (IMR, Philips Healthcare, Cleveland, OH, USA) and filtered back projection (FBP) methods.Materials and methodsWe used non-enhanced porcine liver to simulate the human liver and acquired it 100 times to obtain the average FBP value as the ground-truth. The mean and the standard deviation (“inter-scan SD”) of the pixel values on the 100 image acquisitions were calculated for FBP and for three levels of IMR (L1, L2, and L3). We also calculated the noise power spectrum (NPS) and the normalized NPS for the 100 image acquisitions.ResultsThe spatial SD for the porcine liver parenchyma on these slices was 9.92, 4.37, 3.63, and 2.30 Hounsfield units with FBP, IMR-L1, IMR-L2, and IMR-L3, respectively. The detectability of small faint features was better on single IMR than single FBP images. The inter-scan SD value for IMR-L3 images was 53% larger at the liver edges than at the liver parenchyma; it was only 10% larger on FBP images. Assessment of the normalized NPS showed that the noise on IMR images was comprised primarily of low-frequency components.ConclusionIMR images yield the same structure informations as FBP images and image accuracy is maintained. On level 3 IMR images the image noise is more strongly suppressed than on IMR images of the other levels and on FBP images.     

Real-time MRI and articulatory coordination in speech

This paper describes the real-time MRI technique and its use for the study of speech production. The two major problems, (i) the simultaneous recording of the MR images and the speech signal and (ii) the synchronisation of the images and of the speech signal, are addressed. Measurement accuracy on real-time images is evaluated by comparison with similar measurements on static MR images.     

肝脏超声图象局部分形指数研究

为了研究人体组织超声图象的局部特征,并为进行人体组织定征研究提供新的参数,提出了一个分析超声图象局部分形指数的新方法－局部分形指数小波分析法ＬＦＷＡＭ（ＬｏｃａｌＦｒａｃｔａｌＳｃａｌｅＷａｖｅｌｅｔＡｎａｌｙｓｉｓＭｅｔｈｏｄ）。应用此法研究了人体肝脏组织超声图象分形体的构造规则;进行了局部分形指数的分析。验证了ＬＦＷＡＭ法分析肝脏超声图象局部特性的有效性,得出了局部分形指数更能全面、细致地刻画肝脏组织超声图象分形特征的结论,为进而研究局部病变的识别与图象的分割提供了基础。