Issuu on Google+

International Journal of Computer Science Engineering and Information Technology Research (IJCSEITR) ISSN 2249-6831 Vol. 3, Issue 3, Aug 2013, 85-94 Š TJPRC Pvt. Ltd.

AN IDENTIFICATION OF WHEAT RUST DISEASES IN DIGITAL IMAGES:A REVIEW HITESHWARI SABROL1 & SATISH KUMAR2 1

Department of Computer Science & Applications, G.G.S.C.W, Sector-26, Panjab University, Chandigarh, India

2

Department of Computer Applications, P.U. SSG Regional Centre, Panjab University, Chandigarh, Hoshiarpur, India

ABSTRACT In this review, we are trying to identify the rust diseases of wheat crop in digital images. Although wheat, like all other cereals, is attacked by large number of plant pathogens, the rust fungi are considered to be the most destructive among them. Wheat in India is attack by all three rusts, namely, black or stem rust, brown or leaf rust and yellow or stripe rust. The annual loss due to diseases is more than 50 percent. After collecting all the details, we are trying to identify wheat rust diseases by taking its samples. On the basis of these samples, we train fuzzy-neuron network and by using some image processing or sequential techniques to prevent wheat crop from heavy losses in production of crop and yield losses at early stages of diseases occurrence.

KEYWORDS: Wheat Crop Samples, Fuzzy-Neuron Network, Rust Diseases, Clustering, etc INTRODUCTION The diseases and pets of plants can effects the production of crops and results in yield loss. According to FAO, there are 82 countries that grow wheat [36] globally. Twenty seven, the developing countries sow 100,000 hectares or more of wheat annually [33]. South Asia includes India, Pakistan, Nepal, Bengladesh and South East Afganistan are the developing countries, where wheat is grown. In South India and the mountains area of Nepal, India and Pakistan, wheat is grown throughout the year. The wheat crop is grown at higher elevation allow the rust of the crop to over summer to provide inoculums of reinfection of the maize crop sown from September to December in the plains of India, Nepal and Pakistan [33][34][35]. The most critical factor responsible for establishment of rust is temperature in the plains.

WHEAT DISEASE MANAGEMENT The adoption of wheat diseases management programs will reduce loses in yields and grain quality. The correct identification of disease as early will may help in taking action to prevent losses to produce high yields of good quality grain. The various effective methods of controlling major wheat disease are following. Disease Resistance Growing resistant varieties is an economically and efficient method in disease reduction. All the main diseases are not to be resistant by a single variety. The most common and serious disease-resistant varieties should be selected on the basis of local adaptability and high yield potential. High Quality Seed The use of high quality and certified seed is result in consistent qualitative yields. The selected seed must free of mechanical damage and seed borne disease causing pathogens in the fields and it is the best method to prevent crops from diseases at early season growth.


86

Hiteshwari Sabrol & Satish Kumar

Crop Rotation Crop rotation is another very important method of disease reduction. The continuous wheat production produces yellow spot, roots rots, powdery mildew and head blights etc. To control these diseases, rotations must cover at least six years and the best way to control loses is the use of highly resistance varieties. Failure in controlling seedling blights may result in winterkill of diseased seedling. Seed Treatment Seed treatment helps in producing high yields of good quality grain. Seed or fertilizer treatment can control stripe rust up to four weeks after sowing and suppress it thereafter. The combination of more than one fungicides is necessary to obtain seed protection. Planting Site and Time The pathogens survive on or in crop debris, soil, and volunteer wheat and alternate host plants. Destroy volunteer wheat plants by March as they can preside a green bridge for rust carryon. Some other diseases are not affected by choice of planting site including airborne and insect-transmitted diseases and these include wheat rusts, barley yellow dwarf virus etc. Foliar Fungicides Rusts may occur every year regardless of the precautions taken. If the weather conditions are favourable then these diseases may cause losses of 10 to 30 percent. Rust can be controlled by applying fungicides at proper time and rates. If label recommends and fungicides spray is needed consider crop stage and potential yield loss.

WHEAT RUST DISEASES Black or Stem Rust [Puccinia Graminis Tritci. Eriks. & Henn.] Stem Rust is distributed generally with the wheat crop. In dries areas, the disease develops in epiphytotic form only in moist seasons. This disease probably has caused greater and more spectacular damage than any other disease of the wheat crop. Losses are higher in the spring wheat. This is due apparently to two main factors: (1) the relatively high summer precipitation in the spring wheat areas and (2) the plant growth occurring over a long period of favourable summer conditions. In the plains, it appears late in the season, particularly in areas which are far away from the hills. The first symptom of the disease under field conditions is the appearance of small brownish-colored pustules in the aerial parts of plants, particularly on the lower leaves, the leaf sheaths and the stem. These pustules gradually enlarge and coalesce with each other, forming big lesions of dark-brown color, covered by a membranous epidermis, which ruptures as the pustules enlarge, releasing a fine powdery mass of brown spores. Even at a later stage the remnants of epidermis are seen on the fringes of the well developed pustules. In case of serious attacks, clouds of uredospores are produced. In the later stages of infection dark-colored teleutospores are produced in the pustules which gradually turn black in color, and for this reason the rust is commonly known as the „Black Rust‟ (Figure 1 and Figure 2).The uredospores are echinulate, oval in shape, 25 to 30µ * 17 to 20µ, brown in color and teleutospores are black in color, bi-celled, pedicellate, 40 to 60µ * 15 to 20µ. The infected foliar tissues die earlier than the healthy ones, and in case of severe rust attack the plant size is reduced, tillering is poor and the grains get shrivelled and are light in weight. In extreme cases the grain is not formed at all. The rusted straw becomes dry and brittle. It has also been recorded that the water requirement of rusted plants is much higher than that of healthy ones.


87

An Identification of Wheat Rust Diseases in Digital Images: A Review

Figure 1: Stem Rust Pustules on a Susceptible Wheat Stem

Figure 2: Stem Rust Pustules on a Susceptible Wheat Leaf

Brown Rust or Leaf Rust [Puccinia recondita Rob. Ex Desm.] The brown or leaf rust of wheat generally distributed through the humid and semi-humid wheat-producing areas of the world. The brown or leaf rust suggest, the fungus is mainly restricted to the leaves and produces spores of light brown color on the infected parts (Figure 3 and Figure 4). This stage, however is not restricted to the leaf blade alone in cases of severe infection may be found on leaf sheaths, glumes and awns as well. The uredo-pustules first appear on the upper surface of the leaf blade and are light orange in color. Later these may be found on surface of the leaf through are most common on the upper surface. At first these are sub-epidermal but soon the epidermis ruptures with longitudinal slits and brown powdery uredo-dust is exposed and is easily disseminated by wind currents. The teleuto-stage, on the other hand, is found mostly on the lower surface. The uredo-pustules, as a rule, are never in rows but are scattered irregularly on the leaf surface. There is no discoloration of the surrounding tissue in the early stages of development but later the leaves becomes completely, or at least partially yellow. The uredial size is much smaller than that of black rust, being only 0.5 to 2.5mm. If pustules are few and scattered, these get surrounded by a partial or complete circle of smaller uredia. In case of severe attack the disease leads to premature death of the plant. In severe epiphytotics the plants are reduced to half their normal size and the root system also deteriorates.

Figure 3: Leaf Rust Pustules on a Susceptible Wheat Leaf

Figure 4 : Leaf Rust Infection on a Susceptible Wheat Plant

Yellow or Stripe Rust [Puccinia glumarum. Eriks. & Henn] This rust is easily distinguishable from the two rusts of wheat discussed above by its color. Apparently this rust is limited in its distribution to the areas of relatively cool summer temperatures and humid winters. The rust is known as glume rust, yellow rust as well as stripe rust (Figure 5 and Figure 6).


88

Hiteshwari Sabrol & Satish Kumar

The rust is characterized by small pinhead-like pustules of yellow color, running in streaks, hence the common name „stripe rust‟. In case of mild attack the uredo-pustules are chief found on the leaf blade but in more serious cases they may appear on the sheath, stalk, glumes, and awns and even on grains. The uredospores are produced in such a great abundance that the soil below the wheat plants turns yellow. The teleutopustules appear as broad columns running throughout the length of the leaf blade if the infection is severe. The infected leaves mature earlier than the healthy ones and give a „dried-up‟ appearance to the whole infected field when viewed from a distance. The uredospores, unlike in black rust, round and measure 23 to 35µ*20 to 35µ, with five spines and a hyaline wall.

Figure 5: Yellow Rust Pustules on a Susceptible Wheat Leaf

Figure 6: Yellow Rust Infection of a Susceptible Wheat Plant in the Field

LITERATURE SURVEY The Crop Management System is an expert system to support irrigated wheat and this system is designed to integrate generic task second generation expert systems methodology first developed by Chandrasekaran [31] and the CERES crop simulation methodology pioneered by Ritchie [32]. The system follows various facts of management like planting data selection, water utilization and management pest monitoring, identification, remediation and harvest management [29]. The knowledge based system provides efficient knowledge acquisition, storage, knowledge engineering, processing and proper maintenance of knowledge that can be ultimately used by the diagnostic expert system. KMSCD is based on Knowledge engineering that contains Knowledge acquisition, Knowledge structuring knowledge representation, knowledge verification and validation for Crop Diseases [17]. Rusts are the most important fungal diseases. The climate fluctuations have created a suitable environment for the incidence of various kinds of pests and diseases [4]. The weather conditions are if optimal then the capacity to form new races that can attack previously resistant cultivars, ability to move long distances and potential to develop rapidly. The wheat rust fungi are obligate parasites; that is, they can grow and multiply in nature only on living plant tissue. They are highly specialized in the plant species they parasitize. The complete disease cycle is complicated; two plant hosts (wheat and an alternate host) and several different spore types are required to complete the life cycle [2]. The Support Vector Machine (SVM) based prediction approach will open new vistas in the area of forecasting plant diseases of various crops [24]. [19] This paper measured the infection dynamics (of powdery mildew (Blumeria graminis) and leaf rust (P. recondita) using three high-resolution remote sensing images and a decision tree was constructed to classify the levels of disease severity using mixture tuned matched filtering (MTMF) results and the Normalized Difference Vegetation Index (NDVI). These two indices are used to process of analyzing crop contamination stress sensibility. On the basis of these parameters, a dynamic fuzzy-neural network can build. Three types of remotely sensed data can be generally categorized in identifying


An Identification of Wheat Rust Diseases in Digital Images: A Review

89

and assessing the crop diseases: ground-based, airborne and space borne data but sometime hyper-spectral and coarse spectral resolution are insufficient, inaccurate and decrease the finer identification accuracy of disease plants [5]. The statistical analysis and multilayer perception models can simulate the wheat production through given plantation areas and statistical analysis plays a key role in identifying the most related factor, detecting outliers, determining the general trend of wheat yield. The combination of these two methods provides both meaningful qualitative and accurate quantitative data analysis and forecasting [1]. The remote sensing technology is used to analyse reflected solar radiation from crop canopies to determine disease severity level and assess the health [9]. Another technique called hyper spectral remote sensing for discriminating different fungal infection levels of rice panicles under laboratory conditions [13]. Using spectro-radiometer and a multispectral CCD camera to detect soybean rust and quantify image severity [18]. The fuzzy image based paddy disease diagnosis expert systems using the knowledge, the input and output linguistic variables and their values have been identified [22]. This model is used to create a complete fuzzy image based paddy diagnosis system. The decision system based on fuzzy neural network is proposed that could converge faster than a simple back-propagation network [26] and the fuzzy c-means clustering, fast c-means clustering with random sampling generalised fuzzy c-means, anisotropic mean shift based fuzzy c-means algorithms for color quantisation of images is used in color reduction in images [14]. Another fuzzy rule-based classifier presents a general framework to significantly increase the classification accuracy if compared to the case of using all the available input variables. A large number of rules and input variables can be eliminated from the model without determining the classification accuracy [11]. The extent to which wheat grain color objective measured by video colorimetry, can be used to distinguish kernel type according to wheat class and variety was investigated discriminate analysis were performed based on mean red (R), green (G) and blue (B) pixel reactance features contained by color digital image analysis [30]. The colored based segmentation method that uses k-means clustering technique to partition image into k-clusters [3]. To retrieve the images with in a large collection based on color projections and different mathematical approaches are introduced and applied and images are sub grouped using threshold values [7]. To distinguish most of the cultivars and advanced breeding lines in wheat grains by image processing to remove noises using statistical filters [28]. The systems of born toxicity (i.e. necrosis of leaf tips and margins) have been observed on eucalyptus trees using image analysis techniques [27]. The principal component analysis and image processing methods are used to detect different features of petal color pattern [25] and using pattern recognition based software prototype for rice diseases detection based on the infected images of various rice plants [19]. The technique is developed to differentiate early narrow-leaf wheat from two common weeds (rye grass and brome grass) from their digital images using feature (color, texture and shape) extraction and then principal component analysis to reduce these three descriptors [12]. Another approach of image segmentation based on low-level features including color, texture and spatial information technique for extending the feature space filtering in mean shift algorithm using discrete wavelet frames [23] and image segmentation algorithm is used to separates the weed area from soil background according to the color Eigen value, which is obtained by analyzing the color differences in the three color spaces RGB, rgb and HIS [21]. The methods are used for automatic detection and classification of plant diseases, after the segmentation phase it identify the mostly green colored pixels and these pixels are masked based on specific threshold values that are computed using Otsuâ€&#x;s method, then those mostly green pixels with zero red, green and blue values and pixels on the boundaries of the infected cluster (object) were completely removed. This technique is a robust technique for the detection of plant diseases [10]. The Graph based image segmentation techniques called the directional nearest neighbour graph for texture objects and to reduce computational load [6].


90

Hiteshwari Sabrol & Satish Kumar

The classification and recognition of paddy diseases using mathematics morphology [15] and the back propagation neural network classifiers were designed for classifying the healthy and diseased infected rice leaves in images [16]. The multiple scale neural architecture for recognising colored and texture image is preceded by an explanation of the human visual system, bio-inspired models for image segmentation and ART based neural network in image recognition [8]. After studying these methods and techniques, we can draw some conclusions that various remote sensing and hyper spectral remote sensing techniques are used to detect and predict plant diseases but there are more studies are needed to explore the spectral response characteristics of crops under disease stress of different levels in field and weather conditions. The other image processing, fuzzy neural networks and pattern recognition techniques are used to extract features of crop infected images and healthy crop images to distinguish there features on the basis of texture, color and shape. On the basis of these, wheat rust identification features are following: Black or Stem Rust 

At early stage, small-brownish colored pustules lower leafs and stem. The shape is oval, 25-30 µ * 17-20 µ approx. in size.

While forming big lesions of dark-brown color

Dark colored teleutospores are produced, which turned black in color, 40-60 µ * 15- 20 µ approx. in size.

Brown and Leaf Rust 

At initial stage, on leaf blade, are light orange in color and 0.5 to 2.3mm approx. in size

Next it will appear in partially yellow color.

Later, it will be in brown color.

Yellow and Stripe Rust 

The uredia are linear, citron orange yellow low, usually narrow to form stripes on leaves

Round to ovate in shape and 25-35 µ * 20-35 µ approx. in size

Later, it will in yellow color. The above mentioned features may be varying due to environment conditions and the disease occurs when the

environment conditions are adverse. The main objective of this review is to study the current scenario for the identification of wheat rust using image processing, remote sensing, fuzzy-neural networks and pattern recognition techniques in digital images. The diagnosis of the crop is to be made with incomplete information mostly and if the wheat rust is identify at early stages of rust occurrence then the possibility of right diagnosis at right time will may increases where as other important factors like temperature, soil condition, leaf wetness, humidity etc are also responsible for effecting the crop growth. After the study of various methods and techniques, we are trying to identify wheat rust by taking its sample on regular intervals and to design a fuzzy-neural network for detecting wheat rust at its early stages, which helps in provide correct diagnosis. It may also help in reduce the possibility of heavy loss in later stages of crop growth.

CONCLUSIONS In this review, there are so many methods had been implemented to identify and control the crop diseases using


An Identification of Wheat Rust Diseases in Digital Images: A Review

91

images and after analyzing these methods we are trying to identify wheat rust diseases by taking samples on regular intervals and then train fuzzy-neural network which will analyze all the generated results and observation to identify rust affected and non-affected wheat crop areas in next paper.

REFERENCES 1.

William, W. Guo., and Heru Xue, “An incorporative statistic and neural approach for crop yield modeling and forcasting,” Neural Computing and Applications, © Springer-Verlag, Vol. 21, No. 1, pp 109-117, Feb. 2012.

2.

Stephen N. Wegulo., and Emmanuel Byamukama., “Rust Diseases of Wheat” Plant Dieases Wheat, NebGuide, University of Nebraska–Lincol, Nov 2012.

3.

Ms.Chinki Chandhok., Mrs.Soni Chaturvedi., and Dr.A.A Khurshid., “An approach to image segmentation using k-clustering algorithm,” International Journal of Information Technology (IJIT), Vol. 1, No. 1, August 2012.

4.

Ahmed, M., F.U. Hassan., M. Aslam., and M.A. Aslam., “Physiological attributes based resilience of wheat to climate change,” Int. J. Agric. Biol., 14: 407–412. 2012.

5.

Huang, L.S., J.L. Zhao., D.Y. Zhang., L. Yuan., Y.Y. Dong., and J.C. Zhang., “Identifying and Mapping Stripe Rust in Winter Wheat using Multi-temporal Airborne Hyper spectral Images,” Int. J. Agric. Biol., 14: 697–704, 2012.

6.

Liu Zhao., Hu Dewen., Shen Hui., and Feng Guiyu., “Graph-based image segmentation using directional nearest neighbor graph,” Science China Information Sciences., doi: 10.1007/s11432-012-4706-4, Oct 2012.

7.

R.Venkata Ramana Chary., Dr.D.Rajya Lakshmi., and Dr. K.V.N Sunith., “Feature extraction methods for color image similarity” Advanced Computing: An International Journal ( ACIJ ), Vol.3, No.2, March 2012.

8.

M. Anton-Rodriguez, D., Gonzalez-Ortega, F.J., Diaz-Pernas, M., Martinez-Zarzuela., and J. F.Diez-Higuera., “Color-texture image segmentation and recognition through a biologically-inspired architecture,” Pattern Recognition and Image Analysis, Vol. 22, No. 1, pp. 54–68. © Pleiades Publishing, Ltd., 2012.

9.

Hao Zhang., Hao Hu., Xiao-bin Zhang., Lian-feng Zhu., Ke-feng Zheng., Qian-yu Jin., and Fu-ping Zeng., “Estimation of rice neck blasts severity using spectral on BP-neural network,” Acta Physiologiae Plantarum, © Springer-Verlag, Vol 33, No.6 , pp 2461-2466, Nov 2011.

10. H. Al-Hiary., S. Bani-Ahmad., M. Reyalat., M. Braik., and Z. ALRahamneh., “Fast and Accurate Detection and Classification of Plant Diseases,” International Journal of Computer Applications (0975 – 8887) Vol. 17, No.1, March 2011. 11. Antanas Verikas,Jonas Guzaitis, Adas Gelzinis,Marija Bacauskiene., “A general framework for designing a fuzzy rule-based classifier,” Knowledge and Information Systems, Vol. 29, No. 1, pp 203-221, October 2011. 12. Mahmood R Golzarian., and Ross A Frick., “Classification of images of wheat, ryegrass and brome grass species at early growth stages using principal component analysis,” Plant Methods, © Golzarian and Frick; licensee BioMed Central Ltd., Sep 2011. 13. Zhan-Yu Liu., Hong-Feng Wu., and Jing-Feng Huang., “Application of neural networks to discriminate fungal infection levels in rice panicles using hyperspectral reflectance and principal components analysis,” Computers and Electronics in Agriculture, Vol. 72 Issue 2, Pages 99-106, July 2010.


92

Hiteshwari Sabrol & Satish Kumar

14. Gerald Schaefe., and Huiyu Zhou., “Fuzzy clustering for color reduction in images,” Telecommunication Systems, Vol. 40, No. 1-2, pp 17-25, Feb 2009. 15. G. Anthonys., and N. Wickramarachchf., “An Image Recognition System for Crop Disease Identification of Paddy fields in Sri Lanka ,” Fourth International Conference on Industrial and Information Systems, pp. 28 – 31, Dec 2009. 16. Libo Liu., and Guomin Zhou., “Extraction of the Rice Leaf Disease Image Based on BP Neural Network,” 978-14244-4507-3/2009 IEEE. 17. Savita Kohle, Harvinder S.Saini, Raj Kamal, G.K Gupta, KMSCD: Knowledge Mangement System for Crop Diseases, 978-1-4244-5612-3/09/$26.00 c_2009 IEEE. 18. D. Cui M. Li., Q. Zhang., Y. Zhao., and G. L. Hartman., “Detection of soybean rust using a multispectral image sensor,” Sens. & Instrumen. Food Qual, © Springer Science and Business Media, Vol.3, No.1, pp. 49–56, Jan 2009. 19. Santanu Phadikar., and Jaya Sil., “Rice Disease Identification using Pattern Recognition Techniques,” Proceedings of 11th International Conference on Computer and Information Technology, pp. 420 – 423, 24-27 Dec. 2008. 20. Franke, J. and G. Menz., “Multi-temporal wheat disease detection by multi-spectral remote sensing,” Precis. Agric., Vol. 8, pp. 161–172, 2007. 21. Yajing Liu., Fan Yang., Ruixia Yang., Kejin Jia., and Hongtao Zhang., “Research on segmentation of weed images based on computer vision” Journal of Electronics (China) , Vol. 24, No 2 , pp. 285-288, Mar 2007. 22. S. Abdullah1., A. A. Bakar., N. Mustafa., M. Yusuf., S. Abdullah., and A.R. Hamda., “Fuzzy Knowledge Modelling for Image-based Paddy Disease Diagnosis Expert System,” Proceedings of the International Conference on Electrical Engineering and Informatics Institute Teknologi Bandung, Indonesia June 2007. 23. Mustafa Ozden., and Ediz Polat., “A color image segmentation approach for content-based image retrieval,” Pattern Recognition, Vol. 40, No. 4, pp. 1318–1325., April 2007. 24. Rakesh Kaundal., Amar S Kapoor., and Gajendra PS Raghava., “Machine learning techniques in disease forecasting: a case study on rice blast prediction,” BMC Bioinformatics, © Kaundal et al; licensee BioMed Central Ltd., Nov 2006. 25. Y. Yoshioka., H. Iwata., R. Ohsawa., and S. Ninomiya., “Quantitative evaluation of flower color pattern by image analysis and principal component analysis,” Euphytica, Vol. 139, No. 3, pp 179-186, Feb 2004. 26. Arun D. Kulkarni., and Charles D. Cavanaugh., “Fuzzy Neural Network Models for Classification” Applied Intelligence, Vol. 12 No. 3, pp. 207 – 215, June 2000. 27. A. Poss., S.R. Grattan., C.M. Grieve., and M.C. Shannon., “Characterization of leaf boron injury in salt-stressed Eucalyptus by image,” Plant and Soil, Vol. 206, No. 2, pp 237-245, Sep 1999. 28. H. Utku, H. Koksel , S. Kayhan., “Classification of wheat grains by digital image analysis using statistical filters,” Euphytica, Volume 100, Issue 1-3 , pp 171-178, 1998.


An Identification of Wheat Rust Diseases in Digital Images: A Review

93

29. Kamel A., Schroeder K., Shcklen J., Rafea R., and Salah A., “An integrated wheat crop management system based on generic task knowledge based system and CERES numerical simulation,” Proceeding of International Conference on Expert System for Development, pp. 338-345, Mar 1994. 30. M.R. Neuman., H.D. Sapirstein., E. Shwedyk., and W. Bushuk., “Wheat grain color analysis by digital image processing II. Wheat class discrimination,” Journal of Cereal Science, Volume 10, Issue 3, pp. 183–188, Nov 1989. 31. Chandrasekaran B. “Generic tasks in reasoning: High level building blocks for expert system design”, IEEE Expert, pp. 23-30, March 1986. 32. Ritchie, J.T., D.C. Godwin., and S. Otter-Nacker, “CERES Wheat: A simulation model of wheat growth and development college station”, Texas: Texas A & M University Press, 1986. 33. International Maize and Wheat Improvement Center (CIMMYT). “World Wheat Facts and Trends; Report One: An Analysis of Changes in Production, Consumption, Trade and Prices over the Last Two Decades.” CIMMYT, El Batan, Mexico. International Maize and Wheat Improvement Center (CIMMYT). “CIMMYT Review.” El Batan, Mexico. 1978, 1981, 1987. 34. Saari, E. E., “Wheat disease surveillance and its role on genetic vulnerability”, In “Proceedings of the Workshop on Crop Surveillance” (B. H. Zandstra, ed.), pp. 145–155. East-West Center and University of Hawaii, Honolulu, 1977. 35. Saari, E. E., Wheat in the developing countries from the Atlantic to the Pacific. In integreted “Plant Protection for Agricultural Crops and Forest Trees” (T. Kommedahl, ed.), Proc. Symp. Int. Congr. Plant Prot., 9th, Vol. 2, pp. 437–441. Burgess, Minneapolis, Minnesota, 1981. 36. Food and Agriculture Organization (FAO). “Production Yearbook,” Vol. 35. Food Agric. Organ. U.N., Rome, 1981.



10 identification of wheat full