The profession of Social Work.

Write a 750-1,000-word essay on the profession of Social Work. Include the following:
Trace the historical roots of social work from its early European roots to the present.
Discuss social work’s prospective future.

Sample Answer

Many attempts have been made to give a definite meaning to the word Social Work yet no single definition can be given to it. However, there is concurrence that it refers to a profession that encompasses a variety of profession and professional with correlated values, theory and practice. Social workers provide a scrutiny to any problem be it individual and, or societal, economic or political as such their operation is broad and touches every aspect of human life. The focus of this paper is to unearth the historical background



The rural land mass is something other than being a sustaining sourcing in this day and age. Indian economy is profoundly needy of horticultural efficiency. In this way in field of horticulture, identification of sickness in plants assumes a significant job. To recognize a plant illness in beginning stage, utilization of programmed malady identification strategy is helpful. For example an infection named little leaf malady is a perilous illness found in pine trees in United States. The influenced tree has a hindered development and kicks the bucket inside 6 years. Its effect is found in Alabama, Georgia parts of Southern US. In such situations early recognition could have been productive.

The current technique for plant infection recognition is just unaided eye perception by specialists through which recognizable proof and identification of plant sicknesses is finished. For doing as such, a huge group of specialists just as consistent observing of plant is required, which costs high when we do with huge ranches. Simultaneously, in certain nations, ranchers don't have appropriate offices or even thought that they can contact to specialists. Because of which counseling specialists even cost high just as tedious as well. In such conditions, the recommended method demonstrates to be advantageous in checking enormous fields of yields. Programmed location of the illnesses by simply observing the indications on the plant leaves makes it simpler just as less expensive.

Plant malady distinguishing proof by visual way is increasingly relentless assignment and simultaneously, less precise and should be possible just in constrained zones. While if programmed discovery method is utilized it will take less endeavors, less time and become progressively exact. In plants, some broad ailments seen are dark colored and yellow spots, early and late singe, and others are parasitic, viral and bacterial diseases.A model for grouping the yield ailment is proposed. .The current Deep Convolutional Neural Network, MobileNet Architecture is adjusted for crop illness discovery application. This model perceives 16 classes of leaf pictures that contains both infected and sound leaf. The Deep Convolutional Neural Network sets aside a lot of effort for layer calculation and preparing the whole picture dataset. So as to accomplish quick execution, the framework ought to be parallelized utilizing Graphics Processing Unit by Compute Unified Device Architecture (CUDA) programming.When this model is applied progressively, it takes 22 hours for preparing.

Numerous endeavors have been made to get why and how profound learning gets such great exhibitions. Full comprehension of how to pick basic highlights just as how to productively tune hyper-parameters of models is still a long way from being a reality. As of now, profound learning models need a lot of calculation weight to arrive at cutting edge exhibitions on enormous estimated informational collections in disconnected condition. Adjusting the profound learning idea for exactness horticulture has the test of processing the huge measure of accessible pictures in rapid. In farming field, the robots can be utilized to examine the leaves everywhere throughout the field which are the contributions for preparing. The continuous pictures of leaves will contain commotions and contortions. In this way, there is a requirement for building up a profound learning model for diagnosing the yield maladies which is equipped for performing proficient preparing and parallel figuring in brief time with high precision.


• To analyze the maladies in crops like corn, banana, apple and grape with the assistance of picture grouping utilizing profound learning strategy

• To upgrade the presentation of the framework utilizing parallel registering in Graphics Processing Unit(GPU) by utilizing different classifiers.

• To analyze the productivity of plant illness identification utilizing classifiers, for example, SVM, Decision tree, Extreme Learning Machine classifier.


Profound Learning is another region of Machine Learning research, which has been presented with the target of drawing Machine Learning nearer to one of its unique objectives: Artificial Intelligence.A Convolutional Neural Network (CNN), a prevalent design of Deep learning has been utilized in wide assortment of PC vision applications. CNN is involved at least one convolutional layers and afterward pursued by at least one completely associated layers as standard multi-layered neural system.

Figure 1.1 Convolutional Neural Network

The engineering of a CNN is intended to exploit the 2D structure of an info picture. This is accomplished with nearby associations and tied loads pursued by some type of pooling which brings about interpretation invariant highlights. Another advantage of CNN is that they are simpler to prepare and learn less huge parameters than completely associated systems with a similar number of shrouded units. Transformation of picture into 3 Dimensional volume and arrangements of channels at each layer frames the profound engineering. The utilization of channels diminishes the quantity of parameters to be educated. The idea of retraining the current system of MobileNet is used inthis venture.


A broadly useful illustrations preparing unit (GPGPU) is a designs handling unit (GPU) that performs normal numerical and legitimate counts that is generally performed by the CPU (Central Processing Unit). As a rule, GPU's are utilized to render illustrations to application layer. GPGPU are utilized to do errands that were once performed utilizing superior CPU's, for example, encryption/decoding, bit coin mining and other logical calculation. Designs cards are developed for huge parallel applications which in the end cut down the estimation pace of the more dominant CPU's for some parallel assignments. The GPU comprise of shader centers that are liable for rendering various pixels all the while. This can be go to process various picture or information at the same time.

A profoundly arranged GPU may contain numerous shader centers while a multi-center CPU may very well have 8-10 centers. GPGPU have confronted an expanded concentration since the coming for Direct X 10 with brought together shaders in its shader center determination. GPU organizations like AMD/ATI and Nvidia have ways to deal with GPGPU's with their very own API's. AMD utilizes OpenCL while Nvidia utilizes Compute Unified Device Architecture (CUDA) for programming GPUs.

The modules that can be parallelized are to be recognized. With CUDA programming, model ought to be actualized which has same exactness as CPU however with fast of execution. GPU suits well for Deep learning activities like preparing thousand of pictures, processing many middle layers and extricating enormous arrangement of highlights as referenced in NVIDIA site.

Figure 1.2 CPU and GPU


GPU-quickened registering is the utilization of a designs preparing unit (GPU) together with a CPU to quicken profound learning, investigation, and building applications. Spearheaded in 2007 by NVIDIA, GPU quickening agents presently control vitality proficient server farms in government labs, colleges, ventures, and little and-medium organizations around the globe. They assume a gigantic job in quickening applications in stages going from man-made brainpower to vehicles, automatons, and robots.

GPU-quickened registering offloads process escalated segments of the application to the GPU, while the rest of the code still sudden spikes in demand for the CPU. From a client's viewpoint, applications just run a lot quicker.

An illustrations handling unit (GPU) is a specific electronic circuit intended to quickly control and modify memory to quicken the making of pictures in an edge cushion planned for yield to a presentation gadget. GPUs are utilized in implanted frameworks, cell phones, PCs, workstations, and game consoles.

Present day GPUs are exceptionally proficient at controlling PC illustrations and picture preparing, and their profoundly parallel structure makes them more productive than universally useful CPUs for calculations where the handling of huge squares of information is done in parallel. In a PC, a GPU can be available on a video card, or it very well may be inserted on the motherboard or – in specific CPUs – on the CPU bite the dust.

A basic method to comprehend the contrast between a GPU and a CPU is to think about how they process undertakings. A CPU comprises of a couple of centers advanced for successive sequential preparing while a GPU has a hugely parallel engineering comprising of thousands of littler, progressively proficient centers intended for taking care of numerous errands at the same time. NVIDIA GeForce 940M and GT 610 are the two

1.5.1 NVIDIA GeForce 940M

GPU's utilized in this work. The NVIDIA GeForce 940M is a mid-go DirectX 11-good designs card for PCs revealed in March 2015. It depends on Nvidia's Maxwell engineering (GM108 chip) and fabricated in 28 nm. The 940M offers 384 shader units just as 2 GB of DDR3 memory (64 piece, 2000 MHz powerful). GM108 coordinates the 6th era of the PureVideo HD video motor (VP6), offering a superior unraveling execution for H.264 and MPEG-2 recordings. Obviously, VP6 bolsters all highlights of past ages (4K support, PIP, video encoding by means of NVENC API). Lamentably, HDMI 2.0 is as yet not upheld.

1.5.2 NVIDIA GeForce GT 610

Each PC merits devoted illustrations. Breath life into your interactive media execution with a NVIDIA® GeForce® GT 610 illustrations card. Venture up to NVIDIA devoted illustrations for a quicker, increasingly vivid involvement with your preferred applications—unfailingly. It has the accompanying highlights NVIDIA PureVideo HD Technology, Blu-beam 3D Support, TrueHD and DTS-HD Audio Bitstreaming, Microsoft DirectX 11 Support, NVIDIA CUDA Technology, NVIDIA PhysX Technology, NVIDIA FXAA Technology, NVIDIA Adaptive Vertical Sync, HDMI, Dual-connect DVI and PCI Express 2.0 Support.


Harvest malady identification model is coded utilizing TensorFlow system and Python. TensorFlow is an open source library intended for numerical calculation utilizing information stream charts. Hubs speak to scientific activities and the chart edges speak to the tensors conveyed between th