Fruit Images Research Articles

Fruit and vegetable recognition by fusing colour and texture features of the image using machine learning

Efficient and accurate recognition of fruits and vegetables from the images is one of the major challenges for computers. In this paper, we introduce a framework for the fruit and vegetable recognition problem which takes the images of fruits and vegetables as input and returns its species and variety as output. The input image contains fruit or vegetable of single variety in arbitrary position and in any number. The whole process consists of three steps: 1) background subtraction; 2) feature extraction; 3) training and classification. K-means clustering-based image segmentation is used for background subtraction. We extracted different state-of-art colour and texture features and combined them to achieve more efficient and discriminative feature description. Multi-class support vector machine is used for the training and classification purpose. The experimental results show that the proposed combination scheme of colour and texture features supports accurate fruit and vegetable recognition and performs better than stand-alone colour and texture features.

Secure and Robust Watermarking Using Wavelet Transform and Student t-distribution

In this paper, we propose secure and robust data hiding algorithm based on hiding in random coefficients of Discrete Wavelet Transform and student t-distribution. The watermark image is scrambled to increases its security and robustness. Performance evaluation for robustness and imperceptibility of proposed algorithm has been made using bit error rate (BER) and peak signal to noise ratio (PSNR) value for different watermark and cover images such as Lena, Gibbon and Fruit images. The experimental results demonstrate that the proposed algorithm achieves higher security and robustness against JPEG compression as well as other attacks such as addition of noise, low pass filtering and cropping attacks compared to other existing algorithm using DWT coefficients.

Smart Farming: Pomegranate Disease Detection Using Image Processing

Abstract Crops are being affected by uneven climatic conditions leading to decreased agricultural yield. This affects global agricultural economy. Moreover, condition becomes even worst when the crops are infected by any disease. Also, increasing population burdens farmers to increase yield. This is where modern agricultural techniques and systems are needed to detect and prevent the crops from being effected by different diseases. In this paper, we propose a web based tool that helps farmers for identifying fruit disease by uploading fruit image to the system. The system has an already trained dataset of images for the pomegranate fruit. Input image given by the user undergoes several processing steps to detect the severity of disease by comparing with the trained dataset images. First the image is resized and then its features are extracted on parameters such as color, morphology, and CCV and clustering is done by using k-means algorithm. Next, SVM is used for classification to classify the image as infected or non-infected. An intent search technique is also provided which is very useful to find the user intension. Out of three features extracted we got best results using morphology. Experimental evaluation of the proposed approach is effective and 82% accurate to identify pomegranate disease.

An image segmentation approach for fruit defect detection using k-means clustering and graph-based algorithm

Machine vision has been introduced in variety of industrial applications for fruit processing, allowing the automation of tasks performed so far by human operators. Such an important task is the detection of defects present on fruit peel which helps to grade or to classify fruit quality. Image segmentation is usually the first step in detecting flaws in fruits and its result mainly affects the accuracy of the system. A diversity of methods of automatic segmentation for fruit images has been developed. In this paper, a hybrid algorithm, which is based on split and merge approach, is proposed for an image segmentation that can be used in fruit defect detection. The algorithm firstly uses k-means algorithm to split the original image into regions based on Euclidean color distance in $$L^*a^*b^*$$ L?a?b? space to produce an over-segmentation result. Then, based on a graph representation, a merge procedure using minimum spanning tree is then taken into account to iteratively merge similar regions into new homogenous ones. This combination is an efficient approach to employ the local and global characteristic of intensities in the image. The experiment showed good results in the terms of human observation and in processing time.

Automatic method of fruit object extraction under complex agricultural background for vision system of fruit picking robot

Fruit picking robot is required for agricultural automation for fruit harvest, and vision system is the important and crucial composition of a robot system. An automatic extraction method of fruit object under complex agricultural background for vision system in fruit picking robot is presented in this study. The method is based on an improving Otsu threshold algorithm using a new feature in OHTA color space. Color features are extracted in OHTA color space and then used as an input for the Otsu threshold algorithm which calculates the segmentation threshold value automatically. Four kinds of fruit images are selected to validate the automatic extraction method. The fruit objects are automatically extracted with this method and the outputs are presented in binary images. Numerous of experiments show that the automatic extraction method can extract mature fruit from complex agricultural background and the extraction accuracy is more than 95%. The results indicate an effective fruit object extraction method for vision system of fruit picking robot.

Fruit classification using computer vision and feedforward neural network

Fruit classification is a difficult challenge due to the numerous types of fruits. In order to recognize fruits more accurately, we proposed a hybrid classification method based on fitness-scaled chaotic artificial bee colony (FSCABC) algorithm and feedforward neural network (FNN). First, fruits images were acquired by a digital camera, and then the background of each image were removed by split-and-merge algorithm. We used a square window to capture the fruits, and download the square images to 256×256. Second, the color histogram, texture and shape features of each fruit image were extracted to compose a feature space. Third, principal component analysis was used to reduce the dimensions of the feature space. Finally, the reduced features were sent to the FNN, the weights/biases of which were trained by the FSCABC algorithm. We also used a stratified K-fold cross validation technique to enhance the generation ability of FNN. The experimental results of the 1653 color fruit images from the 18 categories demonstrated that the FSCABC–FNN achieved a classification accuracy of 89.1%. The classification accuracy was higher than Genetic Algorithm–FNN (GA–FNN) with 84.8%, Particle Swarm Optimization–FNN (PSO–FNN) with 87.9%, ABC–FNN with 85.4%, and kernel support vector machine with 88.2%. Therefore, the FSCABC–FNN was seen to be effective in classifying fruits.

THE QUALITY IDENTIFICATION OF FRUITS IN IMAGE PROCESSING USING MATLAB

The ability to identify the fruits based on quality in the food industry which is the most important technology in the realization of automatic fruit sorting machine in order to reduce the work of human and also time consuming. This work presents a hierarchical grading method applied to the Fruits. In this work the identification of good and bad Fruits is focused on the methods using MATLAB. First we extract certain features from t he input fruit image, later using different method like thresholding, segmentation, k-means clustering and thus we get related databases. Comparing several trained databases, we get a specific range for the good and bad fruits. From the proposed range we can identify the good and bad fruits. Thus this paper analysis the good and bad fruits with a very high accuracy successfully using image processing.

North American distribution ofEleocharis mamillata(Cyperaceae) and confusion withE. macrostachyaandE. palustris

Abstract Herein, we provide a detailed map and a list of specimens we have examined of Eleocharis mamillata in North America. No chromosome counts have been published for North America under the name E. mamillata. However, the n = 16 chromosome number reported under E. macrostachya from the Queen Charlotte Islands in British Columbia by Taylor and Mulligan (1968) applies to E. mamillata because the voucher specimens, originally labeled E. macrostachya, are actually E. mamillata. Eleocharis mamillata was confused until recently with E. palustris and E. macrostachya in North America and with E. palustris in Europe. It is readily distinguished from the similar E. palustris, especially by perianth bristle number, stomate morphology, and tubercle always sessile. Descriptions and images of micro-morphological characters of the epidermis of members of the E. palustris complex have been published for Europe but not for North America. We provide descriptions and images of fruits and epidermises of selected speci...

Non-destructive measurement of total soluble solid (TSS) content of dates using near infrared (NIR) imaging

The total soluble solid (TSS) content is one of the important quality attributes that determines the commercial value of dates. At present, TSS content in dates is measured using destructive techniques that are laborious and time consuming. In this study, the potential of a near infrared (NIR) imaging technique to determine TSS content of dates was evaluated. Date samples of three cultivars (Fardh, Khalas and Naghal) at tamar stage (n=400 fruits) and two ripening stages (tamar and khalal) of Khalas cv. (n=400 fruits) were used in this study. The area scan NIR camera which covers the full spectrum range of 900‒1,700 nm was used to capture the images of individual date fruits. Immediately following image acquisition, the TSS content (ºBrix) of each sample was measured using a digital refractometer. The TSS content of dates was in the range of 34‒63° Brix and 14‒37°Brix for tamar and khalal stages, respectively. Similarly, the percentage of NIR reflectance was 12‒38 and 39‒64 for tamar and khalal stages, respectively. A multiple linear regression (MLR) prediction model along with a dummy variable concept was established between measured TSS and NIR reflectance. NIR reflectance was inversely proportional to the TSS content of dates. The correlation coefficient (r) between TSS content and NIR reflectance was 0.62 and 0.98 for three cultivars and two ripening stages, respectively. The high correlation among ripening stages indicates the opportunities of NIR technique for potential applications in influencing ripening stages. However, further work is warranted to determine the correlation between TSS and NIR reflectance in different ripening stages of other cultivars.

Statistical Methods for Quantitatively Detecting Fungal Disease from Fruits' Images

In this paper we have proposed statistical methods for detecting fungal disease and classifying based on disease severity levels. Most fruits diseases are caused by bacteria, fungi, virus, etc of which fungi are responsible for a large number of diseases in fruits. In this study images of fruits, affected by different fungal symptoms are collected and categorized based on disease severity. Statistical features like block wise, gray level co-occurrence matrix (GLCM), gray level runlength matrix (GLRLM) are extracted from these images. The nearest neighbor classifier using Euclidean distance was used to classify images as partially affected, moderately affected, severely affected and normal. The average classification accuracies are 91.37% and 86.715% using GLCM and GLRLM features. The average classification accuracy has increased to 94.085% using block wise features.

Psychophysical models of consumer expectations and colour harmony in the context of juice packaging

AbstractThe aim of this study was to develop psychophysical models that predict the influence of pack colours on consumers' psychological responses of fruit juices, such as visually perceived expectations of freshness, quality, liking, and colour harmony. Two existing colour harmony models derived from experiments involving only uniform colour plaques were tested using the juice packaging experimental data. Both models failed to predict the visual results obtained. Nevertheless, two parameters relevant to chromatic difference and hue difference were somewhat associated with the visual results. This suggested that, among all colour harmony principles for uniform colours, only the equal‐hue and the equal‐chroma principles can be adopted to describe colour harmony of packaging used for juice. This has the implication that the principles of colour harmony may vary according to the context in which the colours are used. A new colour harmony model was developed for juice packaging, and a predictive model of freshness was derived. Both models adopted CIELAB colour attributes of the package colour and the fruit image colour to predict viewers' responses. Expected liking and juice quality can be predicted using the colour harmony model while expected freshness can be predicted using the predictive model of freshness. © 2013 Wiley Periodicals, Inc. Col Res Appl, 40, 157–168, 2015

Mothers, Daughters, Dolls: On Disgust in Elena Ferrante’s La figlia oscura

This article suggests disgust as a framework for approaching the novels of contemporary Italian writer Elena Ferrante. More specifically, it proposes disgust as the operative trope of Ferrante’s figurations of daughterhood, motherhood, and the (pregnant) female body. Ferrante’s novel, La figlia oscura (2006) narrates the experiences of motherhood and femininity through its middle-aged protagonist’s obsession with a young mother, her daughter, and the daughter’s pregnant doll. Images of rotting fruit, insects, oozing wounds, dark gaping mouths, vomit, wombs, worms, snot, sweat, and slime permeate and enliven the text in key narrative moments which always revolve around the doll. Thus, the doll provides us with a case study for examining the workings of disgust in the novel. Read against the backdrop of Ferrante’s own theoretical apparatus, la frantumaglia, and in the context of Western thought on disgust, La figlia oscura can be said to employ the disgusting to problematize motherhood and daughterhood as normative, transparent categories. The article argues that, ultimately, disgust in Ferrante’s texts opens up space for transgression and liberation so that feminine identity becomes slippery and hence can be negotiated across the ‘threshold of repugnance’ that disgust establishes and maintains.

Development of a Hyperspectral Computer Vision System Based on Two Liquid Crystal Tuneable Filters for Fruit Inspection. Application to Detect Citrus Fruits Decay

Hyperspectral systems are characterised by offering the possibility of acquiring a large number of images at different consecutive wavebands. To ensure reliable and repeatable results using this kind of optical sensors, the intensity shown by the objects in the different spectral images must be independent from the differences in sensitivity of the system for the different wavelengths. The spectral efficiency of the acquisition devices and the spectral emission of the lighting system vary across the spectrum and the images, and therefore the results can reproduce these variations if the system is not properly calibrated and corrected. This is particularly complex, when several LCTF devices are used to obtain large spectral ranges. This work presents the development of a hyperspectral system based on two liquid crystal tuneable filters for the acquisition of images of spherical fruits. It also proposes a methodology for acquiring and segmenting images of citrus fruits aimed at detecting decay in citrus fruits that has been capable of correctly classifying 98 % of pixels as rotten or non-rotten and 95 % of fruit.

Shape Oriented Feature Selection for Tomato Plant Identification

Selection of relevant features for classification from a high dimensional data set by keeping their class discriminatory information intact is a classical problem in Machine Learning. The classification power of the features can be measured from the point of view of redundant information and correlations among them. Choosing minimal set of features optimizes time, space complexity related cost and simplifies the classifier design, resulting in better classification accuracy. In this paper, tomato (Solanum Lycopersicum L) leaves and fruiting habits were chosen with a futuristic goal to build a prototype model of leaf & fruit classification. By applying digital image processing techniques, tomato leaf and fruit images were pre-processed and morphological shape based features were computed. Next, supervised filter and wrapper based feature selection techniques were adopted to choose the optimal feature set leading to small within-class variance and large among-class distance which may be of utter importance in building the model for recognition system of the tomato leaf and fruiting habit genre.

Relationship between total solid content and red, green and blue colour intensity of strawberry (Fragaria x ananassa Duch.) fruits

The relationship between total solid contents and red, green, blue (RGB) colour intensity of full spectrum images of strawberry fruits were examined to develop an efficient, non destructive, cost effective and economical method to determine correct maturity stage of strawberry fruits for harvesting. The RGB colour intensity of the full spectrum images of strawberry fruits of cv. Chandler, RGB colour ratio and their respective total solid content (Brix value) at three different colour development stages; fully red coloured (FRC), three-quarter coloured (TQC) and half coloured (HC) were analyzed. Total solid contents of the strawberry fruits were significantly difference among the three colour developmental stages of fruits. The highest average total solid contents (7.70 Brix ) was found in the fully red coloured strawberry fruits while the lowest total solid content (5.30 Brix ) was found in the half coloured fruits. Total solid content of three-quarter coloured fruits was 6.10 Brix. The sum of red, green and blue colour (R+G+B) intensities and the ratio of ((G+B-R)/(R+B+G))2 were strongly correlated (R2=0.95 and 0.89 respectively) with the total solid content of strawberry fruits. This two indices could be used to decide the correct maturity stage of the strawberry fruits. With reducing the values of these indices, increase the TSS of fruits. consequently, the lowest values of indices indicates the highest total solid contents in fruits . DOI: http://dx.doi.org/10.4038/jas.v8i2.5743 The Journal of Agricultural Sciences, 2013, vol.8, no2 p.82-90

An Adaptive Thresholding algorithm of field leaf image

In this paper, we analyze the background and foreground images of jujube leaf, and propose a new Adaptive Thresholding algorithm that can segment single leaves in a leaf image extracted randomly from an online system. We use the OTSU and CANNY operators to segment the area of the target leaf by choosing the thresholds with the Mapping Function, the Shape Identification algorithm and pattern recognition. The optimization process of the algorithm, which includes Mapping Function, the Shape Identification algorithm, morphological methods and logical operations, is designed to precisely obtain the entire leaf edge. This algorithm has an advantage when segmenting complicated leaf images that contain overlapping laminas and have an uneven gray scale in the leaf region itself. Experiments show that this algorithm is both feasible and effective in segmenting jujube leaf images from real-time video systems, and we can obtain clear, smooth, accurate edge images. The algorithm can be used for other kinds of leaf or fruit image segmentation tasks after debugging and improvement.

Application of optical coherence tomography to non-destructively characterise rind breakdown disorder of ‘Nules Clementine’ mandarins

The feasibility of optical coherence tomography (OCT) for imaging histological changes associated with the development of a progressive rind breakdown (RBD) disorder of ‘Nules Clementine’ mandarin (Citrus reticulate Blanco.) was investigated. The investigation utilised fruit with different levels of the disorder, carefully selected from a batch of fruit stored for eight weeks at 8±0.5°C. Images of healthy and RBD-affected intact mandarin fruit were acquired using a Thorlabs OCT system based on a broadband 930nm source. OCT provided high resolution 2D images of fruit rind to a depth of about 1.1mm. Immediate and non-destructive acquisition of images showing histological and microstructural features in intact rind tissues was demonstrated. The oil glands stayed intact in unaffected fruit and gradually collapsed in RBD affected fruit. At advanced stages of the disorder, the collapsed oil glands became increasingly deformed and flattened. The study showed that OCT is a promising technique for immediate, real-time and non-destructive acquisition of images showing histological and microstructural rind features of ‘Nules Clementine’ mandarin fruit.

Is There Fruit In That? Children Are Misled By Television Food Advertisements

Fruit‐flavored food advertisements on television often contain verbal (e.g., words like fruit, fruity, juice) and visual (e.g., images, art) references to fruit despite a lack of real fruit in the product. This study determined how information in television food ads affected children's (4–7y; n=49; 65% male) perceptions of foods’ fruit content. Each child was shown the ads (n=16) in random order and, after each, asked, “Did they use fruit to make this?” The ad groupings were for 100% real fruit foods (n=8) and foods containing 0% real fruit with artificial fruit‐flavors as their only fruit “content” (n=8). Each ad grouping had 4 categories with 2 ads/category: neither fruit images nor verbal references, fruit images, verbal fruit references, and both fruit images and verbal references. All ads were for foods not distributed in the study area to control for familiarity. The presence of fruit images in ads led to the perception that the food contained real fruit, regardless of its true fruit content. The accuracy with which they could identify whether a food was made with fruit was significantly lower for ads of foods containing 0% fruit when the ad had a fruit‐related verbal (p=0.017), visual (p<0.0001), or both references (p<0.0001). Results suggest that children are deceived about foods’ real fruit content when ads for artificially fruit‐flavored foods containing 0% real fruit include fruit images.Grant Funding Source: NJ Agricultural Experiment Station

D3: The Dispersal and Diaspore Database – Baseline data and statistics on seed dispersal

Seed dispersal is hard to measure, and there is still a lack of knowledge about dispersal-related traits of plant species. Therefore, we developed D3, the Dispersal and Diaspore Database (available at www.seed-dispersal.info), which aims at simplifying ecological and evolutionary analyses by providing and integrating various items related to seed dispersal: empirical studies, functional traits, image analyses and ranking indices (quantifying the adaptation to dispersal modes). Currently, the database includes data for more than 5000 taxa and 33 items as well as digital images of diaspores (i.e. the dispersal units), seeds, fruits and infructescences. The included items cover common traits like diaspore mass, size, shape, terminal velocity and seed number per diaspore. Furthermore, we present newly or further developed items like ecomorphological categorizations of the diaspore and fruit as well as information from literature on prevailing dispersal modes. Finally, we introduce several items which are not covered in other databases yet: surface structure and form of the diaspore, the exposure of the diaspores in the infructescence and dispersal rankings. Dispersal rankings allow estimations of how well certain species are adapted to a specific dispersal mode in comparison to a larger species set. They are calculated as the percentile rank of an indicator of species’ dispersal potential in relation to a larger species set. Especially for the new and further developed items we outline the basic concepts in detail, describe the measurement and categorization methods and show how to interpret and integrate these data for single species as well as for larger species sets. Thereby, we calculate baseline statistics of seed dispersal of the Central European flora. We found that diaspores of 72% of the taxa show specializations related to long-distance dispersal, i.e. most often elongated appendages or nutrient-rich tissues. Diaspore masses, sizes and terminal velocities vary over several orders of magnitude and can be approximated by lognormal distributions.

Changing how I feel about the food: experimentally manipulated affective associations with fruits change fruit choice behaviors

Fewer than half of Americans meet current recommendations for fruit and vegetable intake. The behavioral affective associations model posits that feelings and emotions associated with a behavior are a proximal influence on decision making. Cross-sectional evidence supports the model and suggests that affective associations predict fruit and vegetable consumption. The purpose of this study was to test whether a causal relation exists between affective associations about fruits and future fruit consumption behavior, as measured by a snack selection task. Following a baseline assessment of cognitive and affective variables, participants' (N=161) affective associations about fruits were experimentally manipulated with an implicit priming paradigm. Images of fruits were repeatedly paired with positive, negative, or neutral affective stimuli. The key outcome measure was a behavioral choice task in which participants chose between fruit and a granola bar. Participants in the positive prime condition were three times more likely than those in the negative condition to select a piece of fruit over the granola bar alternative in the snack selection task. They were also twice as likely as those in the neutral condition to select fruit. There were no changes in self-reported affective associations or cognitive beliefs. These findings provide further evidence of the implicit and direct influence of affective associations on behavior, suggesting the need to both incorporate the role of affect in health decision making models, as well as the potential utility of intervention strategies targeting affective associations with health-related behaviors.