The initial step for Soil preparation is testing the soil. It involves identifying the soil’s current nutrient levels and the suitable amount of nutrients to be feed to a certain soil based on its fertility and crop demands. The values from the soil test report are being used to categorize a number of key soil parameters, notably Phosphorus, Potassium, Nitrogen, Organic Carbon, Boron, as and soil ph. Irrigation is a type of agriculture that plays an important role in water and soil conservation. Complicated data could be used to maintain irrigation performance and consistency when assessing systems with respect to water, soil, climate, and crop facts. Weeds are plants that is grown where it is not needed. It includes plants that are not intentionally sown. Weeds compete for water, nutrients, light, and space with agricultural plants, lowering crop yields. Weeds can diminish the commercial worth of agricultural regions by lowering the quality of farm products, causing irrigation water loss, and making harvesting machinery harder to run. To control weeds, farmers often spray homogeneous herbicide spraying throughout the field twice or three times during the growth season. However, this method has resulted in the uncontrolled use of large volumes of herbicides, which is harmful to humans, non-target animals, and the environment . Plant diseases can have a devastating influence on food safety, as well as a considerable loss in both the quality and quantity of agricultural goods.As a result, in the field of agricultural information, computerized identification and diagnosis of plant diseases is widely needed. Many approaches for doing this problem have been offered, with deep learning emerging as the preferred method because to its excellent performance. Hence this work focuses on the steps involved in cultivation of crop.

It uses Deep Learning and Machine Learning algorithms to deliver solutions to various challenges faced during cultivation. It mainly focuses on recommending the crops based on weather parameters, stacking pots suggesting the nutrients requirements and specifying the Growing Degree Days. It also helps in identifying the weeds and recommending herbicides for the same. Many insects ruin the crops hence pesticides are recommended based on the insects that are present in the field. And finally cost estimation is very much needed in these recent times. Crisis, uncertainties would result in great loss. Hence forecasting the cost for cultivating a crop is necessary to plan for future uncertain events. This work specifies various costs in cultivation for future years. Crop growth is primarily influenced by the soil’s macro-nutrient and trace mineral content of the soil. Soil being the broad representation of several environmental factors including rainfall, humidity, sunlight, temperature and soil ph. The use of a support vector machine and decision tree algorithm to distinguish the type of crop based on micro-nutrients and meteorological characteristics has been presented as an efficient means of predicting the crop. Three crops where selected such as rice, wheat and sugarcane. Based on certain observations details about micro-nutrients where been obtained. These details where feed into the classifier model that in turn predicted the crop based on the passed values. There are many Machine Learning algorithms that works in a different manner. Hence selecting only two models will not provide the required output. The accuracy score of SVM was greater than decision tree algorithm with a sore of 92%. In this work best out of two algorithms is selected. But there are various algorithms dedicated for classification tasks. There is a need for working on other models such as K Neighbors classifier, Logistic Regression, Ensemble classifiers. These algorithms are indeed applied in proposed research work. The predicts only a crop based on the values entered into the SVM model. Data is most valuable. Hence more information can be obtained apart from using them for prediction. The proposed research work not only recommends the crops and also uses the data to obtain various information that would provide a detailed view about the predicted crops this includes specifying the Growing Degree Days such as heat units, amount of heat needed for the crop growth and the amount of nitrogen, phosphorous and potassium content need to be supplied for the growth per 200 lb. fertilizer. Machine Learning algorithms such as SVM and decision tree classifier was used but in this work Machine Learning algorithms such as Decision Tree, K Nearest Neighbor, Linear Regression model, Neural Network, Naïve Bayes and Support Vector Machine was used for recommending a crop to the user.

It has provided an exposure to other algorithms compared to. Linear Regression model was used to predict the production value against the climatic parameters such as rainfall, temperature and humidity. The scores of all these algorithms were below 90%. This work was just a model implementation using the data set. Web interface needs to be implemented so that even common people can use it efficiently. All the values need to be provided manually for the model to predict the crop. The proposed work helps in extracting temperature and humidity values using Web Scraping. Hence manually entering the values are not needed. The proposed work provides an interactive web interface where the user specifies the average rainfall and soil Ph value. The temperature and humidity details are extracted automatically and feed into the best model that includes 10 algorithms with hyper parameter tuning. The proposed work tends to achieve an accuracy of 95.45% with hyper parameter tuning the algorithms which was not included in. The predicted results along with certain information are displayed in the web interface which makes the user to understand the results more efficiently. Base temperature of a given crop can be used to calculate the GDD Growing Degree Days. The main aim of this study is to come up with easy and mathematically acceptable formulas for calculating GDD’s base temperature. Temperature data for snap beans, sweet corn, and cowpea are used to propose, prove, and test mathematical formulas. These new mathematical formulae, in comparison to earlier approaches, can produce the base temperature quickly and correctly. These formulas can be used to calculate the GDD base temperature for every crop at any developmental stage. This work provides a formula to calculate the GDD for the crops. Hence the formula specified in was applied to the predicted crop to estimate their GDD in the proposed work. Weeds grown along with soybean can be detected using K-means and CNN model. K-means were used for identifying the features of the images and convolutional neural network for was used for classifying the weeds and soybean. It also suggests that accuracy can we improved by fine tuning the CNN model. CNN model provides an efficient way to detect the weeds present among crops. When used along with Kmeans initially the images and its augmentations are clustered and on using CNN model helps to precisely identify the weed. The proposed work uses the pretrained model such as Resnet152V2 hence it has important layers such as skip layer and identity layer. The main goal of these layer is to make sure that the output image is same as the input. This increases the accuracy and the predictions are correct.

Not only predicting the image the proposed model also helps to provide details about the herbicides that can be used which is an additional information for the user. Existing deep learning techniques are used for weed detection. This study provides information of various ML and Deep Learning algorithms that can be used for identifying weeds. It mainly emphasis on pre-trained models. It suggests that pre-trained models as lot of benefits and hence can be used to image classification. It also provides guidance of how to work on datasets and make the datasets efficient for building the models. Many public datasets are available on various platforms that can be used for this purpose. It specifies Image Resizing, data augmentation, image segmentation some of the techniques would bring about accurate classifications and tendency of increasing the accuracy is also more in pre-trained models. Since this study provides directions to perform deep learning techniques the proposed model has opted certain techniques preprocessing steps such as Image Resizing, data augmentation is opted before building the actual deep learning model to predict the weeds. Another algorithm that can be used for identifying weeds in vegetable plantation is the CenterNet. CenterNet is used for weed identification. It includes two stages. In first stage the Bok choy images were collected and detected. In the second stage,grow lights color-index based segmentation were performed on the images collected to identify the weeds present in the dataset. The images were collected from Nanjing, China. The images were augmented to increase the dataset size and images were annotated. CenterNet algorithm was used for both training and testing the images. It is a ground-based weed identification technique. More optimization would lead to better results was suggested. CenterNet algorithm is simple yet there is a need an algorithm that strives to get correct prediction. The proposed work uses Resnet152V2 algorithm that strives to achieve more accuracy since it has special layers such as skip layer and identity layer that tries to get input image as output itself. Hence predictions would be absolutely correct. Hence Resnet152V2 algorithm is selected to obtain accurate prediction and based on the prediction obtain the list of herbicides. Fig. 3.1.

System architecture. Farmers face a challenging task in identifying crop insects since pest infestation destroys a substantial portion of the crop and affects its quality. The use of highly skilled taxonomists to correctly identify insects based on their physical traits is a shortcoming of traditional insect identification. Experiments were conducted using image characteristics and ml algorithms such as neural networks, support vector machine, k-nearest neighbors, naive bayes, and convolutional neural network model to identify twenty-four insects from the Wang and Xie dataset. To increase the performance of the classification models, 9-fold cross-validation was used. The CNN model had the greatest classification rates of 91.5 percent and 90 percent, respectively. The results revealed a considerable improvement in classification accuracy and computational time when compared to state-of-the-art classification algorithms. This work has used basic CNN model for classification as well as the same dataset used by various researchers. Hence the proposed model has used a different dataset called the Pests’ dataset from Kaggle website. This dataset consists of 9 classes of insects. Each image is taken from different locations. This dataset was selected for the proposed model since the model is trained of images about various locations that gives more knowledge for the model to understand the image and distinguish them. The proposed model uses Resnet152V2 model for classification. The Resnet152V2 model is the basic model and top of which Global Average Pooling 2D, Dropouts and more hidden layers are been implemented. This refers to fine tuning the base pre-trained model. This helps in extracting more information and helps in efficient classification. The association between the degree of difficulty in identifying insects and the identification key was investigated in this article. For a collection of 134 insects, the SPIPOLL database was utilized to generate 193 characteristic value pathways. Based on the average IES of all the insects with that of characteristic value was formulated. The CV’s derived IES was then used to generate an estimated IES for each bug, resulting in a ranked list of insects. Finally, the anticipated bug ranking list was compared to the actual bug ranking list. The results showed a significant correlation between the estimated and actual truth IES, indicating that the CV can be used to estimate the IES of SPIPOLL insects. This work has specified of how to consider the features of an image with respect to insects’ dataset. Its main goal is to identify a key that helps in distinguishing the classes. This proposed work contributes in specifying that a key is important for distinguishing the insect classes. Hence the proposed work uses Resnet152V2 algorithm for this very reason. Resnet152V2 is a pre-trained model and it automatically picks the important features rather than manually defining them. The Resnet152V2 base model on addition with Dropouts helps in removing unnecessary hidden layers and selecting the relevant ones is an advantage. Identification of insects does not solve the problem completely. Suggesting Pesticides provides a complete solution.