Visual Search Method Research Articles

Seasonal Diversity of Spiders (Arachnida: Araneae) and Collection Methods in Barpeta District, Assam, India

Barpeta district, Assam, India covers an area of 2754 square kilometers. The Spider specimens were collected by visual search method and the collection methods adopted were Aerial hand collection, Ground hand collection and Beat sheet method. 70 spider species from the different habitats of Barpeta District were documented. The study shows the difference in quantity and quality of spider fauna concerning the seasons and collection method. The spider species were more active in the post-monsoon season and some were inactive during winter. Almost 75% of spider species were collected by the beat sheet method which shows that it is more significant than all other trapping techniques. The spider species richness and spider abundance were significantly affected by seasons, studied by Simpson’s diversity index and Shannon – Weiner’s diversity index. However, further study needs to be carried out for the effective conservation of these species.

AN ACCOUNT OF AMPHIBIAN DIVERSITY AND COMPOSITION IN THREE DIFFERENT HABITAT TYPES IN BAKSA DISTRICT, ASSAM, INDIA

Amphibian assemblages in three different habitat types, rice field; built up habitat and marshland were studied in the Baksa district of Assam, India. Visual encounter survey, opportunistic search and active search methods were used to record amphibians encountered. A total of 1410 individual amphibians were recorded that included 16 species belonging to 11 genera and five families. Among the three habitat types, built up habitat showed maximum species diversity followed by marshland and rice-field. It could be predicted that habitat heterogeneity and architectural complexity are the best predictors of amphibian diversity in the study area. The possible reasons correlating the composition have been discussed. This study indicates the importance of habitat as a resource in the conservation of amphibian species. Further, this happens to be the first report on amphibian assemblage from Baksa district, Assam, India.

Large-Scale Video Retrieval via Deep Local Convolutional Features

In this paper, we study the challenge of image-to-video retrieval, which uses the query image to search relevant frames from a large collection of videos. A novel framework based on convolutional neural networks (CNNs) is proposed to perform large-scale video retrieval with low storage cost and high search efficiency. Our framework consists of the key-frame extraction algorithm and the feature aggregation strategy. Specifically, the key-frame extraction algorithm takes advantage of the clustering idea so that redundant information is removed in video data and storage cost is greatly reduced. The feature aggregation strategy adopts average pooling to encode deep local convolutional features followed by coarse-to-fine retrieval, which allows rapid retrieval in the large-scale video database. The results from extensive experiments on two publicly available datasets demonstrate that the proposed method achieves superior efficiency as well as accuracy over other state-of-the-art visual search methods.

Searching for a face in the crowd: Pitfalls and unexplored possibilities.

Finding a face in a crowd is a real-world analog to visual search, but extending the visual search method to such complex social stimuli is rife with potential pitfalls. We need look no further than the well-cited notion that angry faces "pop out" of crowds to find evidence that stimulus confounds can lead to incorrect inferences. Indeed, long before the recent replication crisis in social psychology, stimulus confounds led to repeated demonstrations of spurious effects that were misattributed to adaptive cognitive design. We will first discuss how researchers refuted these errors with systematic "face in the crowd" experiments. We will then contend that these more careful studies revealed something that may actually be adaptive, but at the level of the signal: Happy facial expressions seem designed to be detected efficiently. We will close by suggesting that participant-level manipulations can be leveraged to reveal strategic shifts in performance in the visual search for complex stimuli such as faces. Because stimulus-level effects are held constant across such manipulations, the technique affords strong inferences about the psychological underpinnings of searching for a face in the crowd.

Species composition of the orb-weaving families Araneidae and Tetragnathidae (Araneae: Araneoidea) in natural habitats in Trinidad, West Indies

ABSTRACTThe biodiversity of some spider groups in the neotropics is poorly documented. The species composition of two orb-weaving spider families, Araneidae and Tetragnathidae in 16 natural habitats was examined. Specimens were collected using visual search and sweep-netting methods. Very few species were found to be abundant in natural habitats while some were common and the majority considered rare. There was also little overlap of species among the localities sampled in each habitat type; overlap ranged from one to three species. Species abundance models showed that natural habitats followed a log series model, indicating that they are characterised by low species richness, the presence of few dominant species, and governed by a single factor. Multi-dimensional scaling analysis showed that four habitats (savanna, mangrove woodland, littoral woodland, and herbaceous swamp) possessed distinct species assemblages. However, the savanna habitat was the most dissimilar, probably due to a greater presence of specialist species.

Short Communication: Herpetofauna diversity, potential ecotourism in Mount Galunggung, West Java, Indonesia

Abstract. Riyanto A, Sulaeman TN, Rachman N, Chaidir DM, Trilaksono W, Farajallah A. 2019. Short Communication: Herpetofauna diversity, potential ecotourism in Mount Galunggung, West Java, Indonesia. Biodiversitas 20: 1173-1179. Here we provide a list of herpetofauna at Mount Galunggung, West Java, Indonesia as potential biodiversity for ecotourism. Data were collected from our observations in March 2013 and November 2018 using visual and acoustic encounter search methods in the three zones (hot springs, craters, and foothills) on the nature tourism destination of the mount. We revealed 35 species of herpetofauna and their local distribution on these zones. Even though the diversity of herpetofauna has not been observed completely, we suggested that the biodiversity data from this zone should be used as the basic capital in turning nature tourism base into ecotourism base, so that it would provide more benefits to biodiversity conservation itself and to the local community.

Artificial Neural Network Based Target Recognition for Marine Search

<p>The key point of marine search and rescue is to find out and recognize the distress objects. At present, the visual search method is usually adopted to detect the ships in distress, and this method can only be used at good sea condition and visibility. In this paper, a new target detection and recognition system is proposed. The parameters of radar transmitter and echo graphics and the invariant moments of radar images are extracted as the system’s recognition features, and the system’s target classifier is based on Artificial Neural Networks (ANN). The developed recognition classifier has been tested using three kinds of target Images, the target’s features are used as the inputs of trained ANN and the outputs of networks are target classification. Sea experimental results show that the proposed method is well-clustering and with high classified accuracy.</p>

Convolutional Neural Network Based Target Recognition for Marine Search

<p>The key point of marine search and rescue is to find out and recognize the distress objects. At present, the visual search method is usually adopted to detect the ships in distress, and this method can only be used at good sea condition and visibility. In this paper, a new target detection and recognition system is proposed. The parameters of radar transmitter and echo graphics and the invariant moments of radar images are extracted as the system’s recognition features, and the system’s target classifier is based on Convolutional Neural Networks (CNN). The developed recognition classifier has been tested using three kinds of target Images, the target’s features are used as the inputs of trained CNN and the outputs of networks are target classification. Sea experimental results show that the proposed method is well-clustering and with high classified accuracy.</p>

Freshwater mussels in an urban watershed: Impacts of anthropogenic inputs and habitat alterations on populations

The substantial increase in urbanization worldwide has resulted in higher emissions of wastewater to riverine systems near urban centers, which often impairs aquatic populations and communities. This study examined the effect of urbanization on freshwater mussel populations, including Species at Risk in two rivers receiving wastewater. The influence of anthropogenic activities was assessed in a watershed in the Laurentian Great Lakes basin, one that historically supported one of the most diverse mussel faunas in Canada. In the Grand River (ON), four sites along a 60km reach spanning from an upstream reference site to an urban-impacted downstream area were examined. In the Speed River, mussel populations at six sites along a 10km reach, selected to bracket specific anthropogenic inputs and structures were assessed. A semi-quantitative visual search method revealed that catch per unit effort in the Grand River declined by >60% from the upstream reference site to the area downstream of an urban center. The size (length) frequency distribution of the most abundant species, Lasmigona costata, was significantly (p≤0.008) different upstream of the majority of urban inputs (45–130mm) compared to downstream of the cities (85–115mm). In the Speed River, impoundments and wastewater treatment plants (WWTP) reduced both the diversity and catch per effort. Most striking were 84 and 95% changes in the number of mussels found on either side of two impoundments, and a 98% drop in mussels immediately downstream of a WWTP outfall. These population level effects of decreased abundance and underrepresentation of smaller mussels downstream of the urban area correspond to previously documented impacts at the biochemical and whole organism level of biological organization in wild mussels at this location. Our results demonstrate that poor water quality and physical barriers in urban environments continue to impair susceptible populations and communities of aquatic animals.

Diversity and equitability of insect pests and natural enemies in brinjal field treated with insecticide

Experiment was conducted during winter season to study the diversity and equitability of insect pest species and natural enemies in insecticide treated brinjal fields. Highest number of insect pests were recorded in` Tracer 45 SC, Bactoil (Bt), Nimbicidene 0.03 EC and lowest was in Necstar-50 EC and Proclaim-5 SG treated plots. The highest total abundance of insect pest was recorded in the plots treated with Bactoil and Tracer-45 SC and lowest total abundance was in Helicide (HNPV), Proclaim-5 SG and Booster-10 EC treated plots. The diversity index and equitability of insect pest species were highest in the plots treated with Nimbicidene 0.03 EC and Bactoil in visual search and sweep net methods while Bactoil and Booster 10 EC in pitfall trap method. However, lowest diversity index and equitability were obtained from the plots treated with Booster 10 EC, Proclaim-5 SG, Necstar-50 EC, Tracer-45 SC in visual search and sweep net methods but also in plots treated with Nimbicidene 0.03 EC in pitfall trap method. In case of natural enemies, the highest number of families were recorded in Tracer-45 SC, Nimbicidine 0.03 EC and Bactoil treated plots while lowest was in Helicide, Booster 10 EC, Proclaim-5 SG and Necstar-50 EC treated plots. The highest total abundance of natural enemy was recorded in the plot treated with Bactoil and Tracer-45 SC while lowest abundance was in the plot treated with Necstar-50 EC and Boster-10 EC. The diversity index and equitability of natural enemies were the highest in the plots treated with Proclaim-5 SG, Bactoil, Helicide and Necstar-50 EC in visual search, sweep net method and pitfall trap method while lowest was in Booster 10 EC, Tracer-45 SC treated plots in visual search method, Booster 10 EC and Nimbicidene 0.03 EC treated plots in sweep net method, Necstar-50 EC and Nimbicidene 0.03 EC treated plots in pitfall trap method. Bactoil and Tracer-45 SC were relatively safe for natural enemies and therefore would be fit well into integrated pest management (IPM) against BSFB of brinjal crop.Jahangirnagar University J. Biol. Sci. 4(1): 71-80, 2015 (June)

Abundance, species richness and diversity of the orb-weaving spider families Araneidae, Nephilidae and Tetragnathidae in natural habitats in Trinidad, West Indies

ABSTRACTThis study examined the biodiversity of orb-weaving spiders in natural habitats in Trinidad, a continental island with a northern South American fauna. Forty-six localities were sampled, classified into seven formations (representing vegetation structure) and 16 habitat types (based on plant species composition), using visual search and sweep-netting methods, resulting in a total of 1225 individuals of 57 species. Abundance and species richness were significantly related to formation but not habitat type, while both factors influenced species diversity, evenness and dominance. Gamma (γ) diversity had a larger component of β than α diversity did, indicating that between-habitat differences are a substantial part of regional diversity in orb-weaving spiders in Trinidad.

Estimating Spring Salamander Detection Probability Using Multiple Methods

Abstract Many studies suffer from imperfect detection probability, i.e., species are not detected when individuals may be present. In occupancy studies, detection probability is often treated as a nuisance variable. When used as a primary variable of interest, detection probability can be examined as a function of sampling covariates with the goal of maximizing the probability of encountering target species. Efforts to determine which methods maximize detection probability will benefit monitoring programs, particularly for species that are difficult to detect. We used three sampling methods, leaf litter bag (LLB) surveys, visual encounter surveys (VES), and flip and search (FS) methods to detect larval Spring Salamanders (Gyrinophilus porphyriticus). We estimated occasion-specific estimates of detection and used an analysis of variance to determine if detection probability varied among sampling methods. We found the FS method yielded higher detection estimates than did the LLB and VES. In addition, occupa...

Evaluating color vision deficiency daltonization methods using a behavioral visual-search method

Daltonization methods are used to automatically improve color images for color-deficient people. A comparison of different daltonization methods, however, is still left undone. We propose a visual-search method to evaluate daltonization methods by assessing behavioral performances of the attentional mechanism through the analysis of accuracy and response time data. Firstly, we show that the visual-search methodology can indeed be used to evaluate daltonization methods. Secondly, we argue that a combination of natural images and Ishihara images is needed to highlight differences between the daltonization methods. Our results indicate that the investigated daltonization methods can be ranked from highest to lowest as following: Firstly, the method proposed by Kotera; secondly, the method proposed by Fidaner et al.; thirdly, the method proposed by Huang et al.; and lastly the method proposed by Kuhn et al.

Occurrence of anurans in brazilian caves

Brazil has the greatest diversity of anurans and also one of the greatest speleological patrimonies in the world. However, informations about anurans in Brazilian caves including different biomes and lithologies are scarce. This study sampled 223 caves divided into different biomes (Amazon, Atlantic Forest, Caatinga, Cerrado and transition area) and lithologies (Conglomerate, Granite, Iron-ore, Limestone, Marble, Quartzite, and Sandstone) distributed in eleven Brazilian states. To determine the anuran composition (presence/absence), a single sampling event was conducted in each cave by a team of three researchers in the period 1999−2011, following acoustic and visual search methods. We recorded 54 species distributed 18 genera and 11 families. The caves in the Amazon biome had the highest number of species, followed by caves present in the Cerrado, Caatinga, transition area (Atlantic Forest and Cerrado) and the Atlantic Forest. The caves in the Iron-ore lithology had the highest number of species, followed by the Limestone, Sandstone, Quartzite, Granite, Marble and Conglomerate caves. The anurans proved to be very diverse in Brazilian caves, with this high species richness related to the large amount of biomes and lithologies sampled. The family Leiuperidae had the highest richness and the species Scinax fuscovarius the highest frequency of occurrence in the caves. Also recorded were tadpoles and immature forms inside caves suggesting that not all the species are accidental, and that some species may be using these environments for shelter, protection, food and, even reproduction.

On-the-fly learning for visual search of large-scale image and video datasets.

The objective of this work is to visually search large-scale video datasets for semantic entities specified by a text query. The paradigm we explore is constructing visual models for such semantic entities on-the-fly, i.e. at run time, by using an image search engine to source visual training data for the text query. The approach combines fast and accurate learning and retrieval, and enables videos to be returned within seconds of specifying a query. We describe three classes of queries, each with its associated visual search method: object instances (using a bag of visual words approach for matching); object categories (using a discriminative classifier for ranking key frames); and faces (using a discriminative classifier for ranking face tracks). We discuss the features suitable for each class of query, for example Fisher vectors or features derived from convolutional neural networks (CNNs), and how these choices impact on the trade-off between three important performance measures for a real-time system of this kind, namely: (1) accuracy, (2) memory footprint, and (3) speed. We also discuss and compare a number of important implementation issues, such as how to remove ‘outliers’ in the downloaded images efficiently, and how to best obtain a single descriptor for a face track. We also sketch the architecture of the real-time on-the-fly system. Quantitative results are given on a number of large-scale image and video benchmarks (e.g. TRECVID INS, MIRFLICKR-1M), and we further demonstrate the performance and real-world applicability of our methods over a dataset sourced from 10,000 h of unedited footage from BBC News, comprising 5M+ key frames.

Top-down Spatial Attention for Visual Search: Novelty Detection-Tracking Using Spatial Memory with a Mobile Robot

Assistive robotics technologies have been growing impact on at-home monitoring services to support daily life. One of the main research fields is to develop an autonomous mobile robot with the tasks detection, tracking, observation and analysis of the subject of interest in the indoor environment. The main challenges in such daily monitoring application, thus in visual search, are that the robot should track the subject successfully in several severe varying conditions. Recent color and depth image based visual search methods can help to handle part of the problems, such as changing illumination, occlusion, and etc. but these methods generally use large amount of training data by checking the whole scene with high redundancy to find the region of interest. Therefore, inspired by the idea that spatial memory can reveal novelty regions for finding the attention points as in Human Visual System (HVS), we proposed a simple and novel algorithm that integrates Kinect and Lidar(Light Detection And Ranging) sensor data to detect and track novelties using the environment map of the robot as a top-down approach without the necessity of large amount of training data. Then, novelty detection and tracking is achieved based on space based saliency map representing the novelty on the scene. Experimental results demonstrated that the proposed visual attention based scene analysis can handle various conditions stated and achieve high accuracy of novelty detection and tracking.

Long-term monitoring of the introduced emerald ash borer (Coleoptera: Buprestidae) egg parasitoid, Oobius agrili (Hymenoptera: Encyrtidae), in Michigan, USA and evaluation of a newly developed monitoring technique

Emerald ash borer (EAB), Agrilus planipennis Fairmaire (Coleoptera: Buprestidae), is a serious invasive pest of ash trees (Fraxinus spp.) in North America from China. The egg parasitoid Oobius agrili Zhang and Huang (Hymenoptera: Encyrtidae) was introduced from China as a biological control agent for this pest in Michigan and throughout the infested area of the United States. A critical component of any biological control program is post-release monitoring and evaluation; however, because of the small size and cryptic nature of O. agrili, evaluation of its impact is difficult. We compared two methods for measuring parasitism of emerald ash borer eggs: (1) timed visual searches of bark on standing ash trees and (2) bark collection, sifting, and sorting. Both methods were carried out in paired parasitoid-release and control plots, the visual search method over a six-year period (2008–2013) and the more recently developed bark-collection and sifting method for 2years (2012–2013). The visual search method found parasitism in release plots remained low (0.7–4.2%) in samples taken from 2008 to 2012 and reached 10.6% in 2013. In comparison, the bark-sifting method found that rates of egg parasitism were considerably higher in release plots, 21.8% and 18.9% for samples taken in 2012 and 2013, respectively. These findings indicate that the population-level impact of O. agrili is increasing and may be an important source of mortality for EAB populations. We recommend the bark-collection and sifting method as the more effective method to recover parasitoids and estimate parasitism rates of O. agrili.

Comparing the Precision of Information Retrieval of MeSH-Controlled Vocabulary Search Method and a Visual Method in the Medline Medical Database.

Background:Medline is one of the most important databases in the biomedical field. One of the most important hosts for Medline is Elton B. Stephens CO. (EBSCO), which has presented different search methods that can be used based on the needs of the users. Visual search and MeSH-controlled search methods are among the most common methods. The goal of this research was to compare the precision of the retrieved sources in the EBSCO Medline base using MeSH-controlled and visual search methods.Methods:This research was a semi-empirical study. By holding training workshops, 70 students of higher education in different educational departments of Kashan University of Medical Sciences were taught MeSH-Controlled and visual search methods in 2012. Then, the precision of 300 searches made by these students was calculated based on Best Precision, Useful Precision, and Objective Precision formulas and analyzed in SPSS software using the independent sample T Test, and three precisions obtained with the three precision formulas were studied for the two search methods.Results:The mean precision of the visual method was greater than that of the MeSH-Controlled search for all three types of precision, i.e. Best Precision, Useful Precision, and Objective Precision, and their mean precisions were significantly different (P <0.001). Sixty-five percent of the researchers indicated that, although the visual method was better than the controlled method, the control of keywords in the controlled method resulted in finding more proper keywords for the searches. Fifty-three percent of the participants in the research also mentioned that the use of the combination of the two methods produced better results.Conclusion:For users, it is more appropriate to use a natural, language-based method, such as the visual method, in the EBSCO Medline host than to use the controlled method, which requires users to use special keywords. The potential reason for their preference was that the visual method allowed them more freedom of action.

Personal Clothing Retrieval on Photo Collections by Color and Attributes

Automatic personal clothing retrieval on photo collections, i.e., searching the same clothes worn by the same person, is not a trivial problem as photos are usually taken under completely uncontrolled realistic imaging conditions. Typically, the captured clothing images have large variations due to geometric deformation, occlusion, cluttered background, and photometric variability from illumination and viewpoint, which pose significant challenges to text-based or reranking-based visual search methods. In this paper, a novel framework is presented to tackle these issues by leveraging low-level features (e.g., color) and high-level features (attributes) of clothing. First, a content-based image retrieval (CBIR) approach based on the bag-of-visual-words (BOW) model is developed as our baseline system, in which a codebook is constructed from extracted dominant color patches. A reranking approach is then proposed to improve search quality by exploiting clothing attributes, including the type of clothing, sleeves, patterns, etc. Compared to low-level features, the attributes have better robustness to clothing variations, and carry semantic meanings as high-level image representations. Different visual attribute detectors are learned from large amounts of training data to extract the corresponding attributes. The construction of codebook and building of attribute classifiers are conducted offline, which leads to fast online search performance. Extensive experiments on photo collections show that the reranking algorithm based on attribute learning significantly improves retrieval performance in combination with the proposed baseline. Even our color-based baseline alone outperforms the previous CBIR-based search approaches. The experiments also demonstrate that our approach is robust to large variations of images taken in unconstrained environment.

Monitoring Epacris muelleri on unreachable cliffs in the Western Blue Mountains, Australia

SummaryAccurate, repeatable estimates of population densities are often desired for vegetation monitoring. However, conventional transect and quadrat field sampling methods are not always applicable to plants such as the rare shrub Epacris muelleri Sond., growing on largely inaccessible cliffs and rock faces. E. muelleri is an ericaceous shrub restricted to the Blue Mountains region in New South Wales, which is to be monitored to detect potential effects of underground coal mining. In this manuscript, we evaluated observer error associated with density estimates to assess suitability of applying the timed‐meander method to this species. The results indicate that a visual search method using binoculars can generate repeatable results among observers with very different experience levels. However, there is a large margin of error in estimating density when there are many plants growing in close association or overlapping on a cliff. Nonetheless, with a strict set of protocols and further evaluation, this method shows promise as a rapid yet robust method for carrying out repeatable surveys for quantifying changes in the population.