当前位置：文档库 › From Bag-of-Visual-Words to Bag-of-Visual-Phrases Using__ n-Grams

From Bag-of-Visual-Words to Bag-of-Visual-Phrases Using__ n-Grams

From Bag-of-Visual-Words to Bag-of-Visual-Phrases using n-Grams

Glauco V.Pedrosa and Agma J.M.Traina

Instituto de Ciências Matemáticas e de Computa??o-ICMC

Universidade de S?o Paulo-USP

{gpedrosa,agma}@https://www.wendangku.net/doc/4316772984.html,p.br

Abstract—The Bag-of-Visual-Words has emerged as an effective modeling approach to represent local image features.It describes local image features by assigning them a visual word according to a visual dictionary.The image representation is given by the frequency of each visual word in the image,as a similar representation used in textual documents.In this paper, we present a novel approach building a high-level description using a group of words(phrases)for representing an image. We introduce the use of n-grams for image representation, based on the idea of“Bag-of-Visual-Phrases”.In the?eld of computational linguistics,an n-gram is a phrase formed by a sequence of n-consecutive words.As analogy,we represent an image by a combination of n-consecutive visual words.We made representative experiments using three public benchmark databases of textures and nature scenes and two medical databases to demonstrate an area that can bene?t from the proposed technique.Our proposed Bag-of-Visual-Phrases approach improved up to44%the retrieval precision and up to33%the classi?cation rate compared to the traditional Bag-of-Visual-Words,being a valuable asset for content-based image retrieval and image classi?cation.

Keywords-Image description;CBIR;Bag-of-features;SIFT;

I.I NTRODUCTION

Image representation plays an essential role in image categorization and retrieval applications.Research in this area has advanced in order to obtain methods that capture the semantics of the image,extracting features perceptually ef?cient and compact[1].This fact should be taken into consideration to make a CBIR(Content-Based Image Retrieval)system closer to the users’expectation.

One of the state of art technique for image representation is the Bag-of-Visual-Words[2],[3],[4],[5],[6],[7],[8], also known as Bag-of-Features or Bag-of-Keypoints.This approach describes an image using a dictionary composed of different visual words.Visual words are local image patterns,which concentrate relevant semantic information about the image.As illustrated in Figure1,after extracting the local image features,each feature is assigned to its nearest visual word according to a visual dictionary.The traditional image representation,employed by the Bag-of-Visual-Words approach,is the number of occurrences of each visual word contained in the image,as an analogy to the Bag-of-Words representation used for textual information retrieval.

A more powerful description can be obtained by grouping words[9],[10],[11],[12],once this approach can encode the arrangement between the visual words in the image space. In fact,aggregating spatial information in visual words is a promising approach for image description,because the appearance of the visual words can change profoundly when they participate in relations.Spatial information reaches a high-level semantic characterization and leads to a more meaningful image representation.

The goal of this work is to represent an image taking into consideration the relationship between the visual words, instead of considering the image as a set of isolated words. In this paper,we introduce the idea of using n-grams for generating visual phrases.The use of n-grams is an ef?cient model already used in natural language processing to represent a textual document[13].The words are modeled such that each n-gram is composed of n sequential words.For example, the1-gram(unigram)representation is composed of isolated words,such as{intelligence,arti?cial,computer,vision, medical,systems}.By analogy,the1-gram representation is the traditional Bag-of-Visual-Words approach.On the other hand,the2-gram(bigram)is represented by two sequences of words,such as{arti?cial intelligence,computer vision, medical systems,arti?cial systems},enriching the semantic and yielding a more complete representation.The core advantages of using n-gram models are the simplicity and the ability to scale up the content representation very effectively by simply increasing n.

We performed representative evaluations in several different databases.We evaluated our proposed method in benchmark databases of the image classi?cation and retrieval area, and also in medical databases,which have shown to be an application area that can bene?t from the proposed technique, since it adds more semantics to the image description. Comparative results,with respect to the traditional Bag-of-Visual-Words approach,show that the use of n-grams is a promising descriptor to achieve a more robust image characterization and make a CBIR system closer to users’expectation.

The rest of the paper is organized as follows.Section2 gives the formal de?nitions and the background needed to follow the work,such as the motivation for the proposed work. Section3explains the method to represent an image in n-gram visual words.Section4presents the experimental analysis and Section5gives the conclusions of this paper.

Figure1.Scheme to represent an image in Bag-of-Visual-Word and Bag-of-Visual-Phrases.

II.B ACKGROUND AND MOTIVATION

The Bag-of-Visual-Words(BoVW)approach is a technique used to model the local image features.These local features are described by an unordered set of keypoints,where each keypoint describes representative local image features.The goal is to quantize these features using a visual dictionary. The main idea of using visual dictionaries is to consider that the image visual patterns are similar to textual words present in textual documents.Therefore,an image is composed of visual words as a textual document is composed of textual words. Clustering is a common method in the literature for learning a visual dictionary.A dictionary can be built by clustering the local features detected in a set of training images from the database,such as schematized in Figure2.Formally,let P={p1,p2,...,p z}be the local features detected in a subset of the database image.The visual dictionary is given by a division of P into k distinct clustersπk={C1,C2,...,C k}, so that C1∪C2∪...∪C k=P,C i=?,and C i∩C j=?for i=j and i,j=1...k.

A visual word w i is the centroid of cluster C i.When a new image arrives,its features are extracted and assigned to the nearest visual word w i,for i=1...k,where k is the number of words in the visual dictionary.The image representation employing the BoVW approach is simply the normalized histogram of the quantized visual words detected in the image.

Spatial information is a very important feature for the characterization of images and objects,because the appearance of objects can deeply change when they participate in relations.One of the?rst work to attempt encoding geometric information with Bag-of-Visual-Words is the spatial pyramid[14],which splits the image into hierarchical cells and computes bag-of-visual-words for each cell,concatenating the results at the end.However,it is crucial that the image characterization does not depend on the placement of the image,because the features should be invariant to geometric transformations.Other works employ correlograms of visual words[15]and image splitting by linear and circular projections[16].The method proposed in [17]encode spatial-relationship information of visual words using image space partitions to count the occurrences of the visual words in relation to the other visual words positions. It seems plausible that grouping words might be applied successfully for enriching the BoVW representation[12],once a high semantic information level is reached.The bene?ts of using group of words have been proven to boost local feature matching[11].Previous works employ phrases to model the co-occurrences of the words in local neighborhoods,using methods to encode the spatial layouts with a grid-dependent size[18].Besides that,a big problem of these approaches is that the number of phrases can exponentially grow according to the number of words in a phrase.Thus,it is necessary to select a subset from the entire phrase set.Sophisticated mining or learning algorithms have been proposed for this selection [9],[10],but it may still be risky to discard a large portion of phrases,because some of which may be representative ones for the images.

Instead of using a dictionary formed by a huge list of phrases with different number of words,why do not use a dictionary formed by phrases with a limited but representative number of words?A representation that utilizes phrases with a limited sequence of n words is denominated n-gram.In the

Figure 2.Process used to generate a visual dictionary.Initially,a keypoint detector is applied in a set of training images to detect representative local image points.The detected keypoints are represented by a descriptor that summarizes the information about the region around each keypoint.A dictionary can be built by clustering the keypoints detected.

?elds of computational linguistics and probability,an n -gram is a contiguous sequence of n items from a given sequence of text or speech [13].An n -gram could be any combination of letters.However,the items in question can be phonemes,syllables,letters,words or pairs according to the application.How useful the n -gram representation could be for image description based on the idea of Bag-of-Visual-Phrases?The idea of using n -grams in vision is quite similar to a number of previous works that combine visual words with spatial information.For example,in [12]triplets of visual words are used.The work in [19]use 2-grams ("doublets")from spatially neighboring word pairs.In [20],the authors use "higher order features"(i.e.BoVW +2-grams,3-grams etc),but instead of considering all possible n -grams,they perform feature selection to only pick relevant n -grams.In [21]the authors propose to do joint clustering of nearest feature pairs.In this paper,we propose a slightly different measure of similarity considering n -grams and BoVW,and a simple and different procedure to extract the n -grams from the image compared to [22]and [23].The details of the proposed technique is presented in the next section.

III.T HE PROPOSED METHOD

The proposed method describes an image as a Bag-of-Visual-Phrases,where the Visual Phrases are n -grams extracted from the image.In our proposed model,the image representation is the frequency that each n -gram appears in the image.In this section we explain how to extract the n -grams from an image and how to represent an image using Bag-of-Visual-Phrases.

A.Extracting the n -grams from an image

In text mining,a unigram representation can be obtained by placing a small window over the text,such that we only look at one word at a time.In a similar way,a bigram can be thought of as a window that shows two words at a time and moving this window to the right,one word at a time,in a stepwise manner.This procedure is the same for extracting n -grams from a text with n greater than two.To take this “window”analogy to our image representation problem,we could say that all visual n -grams of an image can be generated by placing a region over each visual word,such that we only look at the n -nearest visual words at time.

Formally,let P ={p 1,p 2,...,p m }be the local features detected in an image.Each local feature p i is represented by

its coordinates (x i ,y i )in the image and it is assigned to a visual word w j .The n -Nearest Visual Word (n NVW)of a local feature p i is given by:

n NVW (p i ,n )={P ?P ||P |=n,?p j ∈P ,

?p r ∈[P ?P ]:d (p i ,p j )

where:

d (p i ,p j )=

(x i ?x j )2+(y i ?y j )2

(2)The next de?nition is needed to give the main building block of our proposed technique.

De?nition 1.The proposed image representation in n -grams is given by the occurrences of the n -Nearest Visual Words considering each local feature p i detected in an image P:

n -gram (P )={n NVW (p i ,n )},for i =1...m.

(3)

where m is the number of keypoints detected in the image.To illustrate,Figure 3shows an example of extracting the 2-grams from an image.For each local point,we look at its closest neighbor point.There is no need to specify a threshold distance,since we get the nearest point.We consider the pair as a 2-gram phrase.In a similar way,to extract the n -grams from an image,with n >2,we just need to increase the number of nearest points.

B.Modeling the image in Bag-of-Visual-Phrases using n -grams

After extracting the n -grams from an image,each n -gram is treated as a visual phrase.To model an image in Bag-of-Visual-Phrases,the next step is to count the number of visual phrases according to a dictionary of visual phrases.This dictionary of visual phrases is given by the all possible combinations of n -grams.However,the size of the dictionary can be exponential with respect to the number of the visual words in the vocabulary.For example,considering 100different visual words,the number of all possible combinations of 2-grams is 1002.

To reduce the size of the dictionary of visual phrases we do not consider n -grams with repeated visual words,this means,an n -gram with n similar consecutive visual words.Additionally,we consider inverted n -gram as the same visual phrase.Inverted visual phrases consists of phrases with the

Figure3.Example of the process used to extract2-grams from an image.For each keypoint we look at its closest neighbor point.To extract n-grams from

an image,with n>2,we just need to increase the number of nearest

points.

(a)(b)

Figure4.Representation of the image in?g.3by:(a)Bag-of-Visual-Words,(b)Bag-of-Visual-Phrase considering2-gram phrases and the proposed dictionary.

same visual words,where the visual words appear in different orders.With these two restrictions the dictionary of visual phrases can be reduced to50%or more.

To illustrate,the2-gram phrases generated by a dictionary formed by the words{w1,w2,w3}is {{w1,w2},{w1,w3},{w2,w3}}.Figure4shows the representation of the image from Figure3considering the traditional Bag-of-Visual-Words and the proposed method of Bag-of-Visual-Phrases using2-gram phrases.C.Calculating the images’distances

To measure the dissimilarity(distance)between two images A and B,we take advantage of the distance between the histograms of Bag-of-Visual-Words and Bag-of-Visual-Phrases of both images.

Let h A and H A be the normalized histograms of Bag-of-Visual-Words and Bag-of-Visual-Phrases of image A, respectively,and h B and H B of image B.The distance used to measure the dissimilarity between the images A and B is given by:

Distance(A,B)=||h A?h B||1+||H A?H B||1(4) where||.||1is the L1distance.

Two images A and B are considered similar when Distance(A,B)→0.

IV.E XPERIMENTAL RESULTS

In this section,we report representative experimental results performed to evaluate the effectiveness of the image representation technique proposed in this work.We compared our method with the traditional Bag-of-Visual-Words representation,which has only the frequency of occurrence of the words in the image.The Bag-of-Visual-Words is by analogy the1-gram representation.

We seek a vocabulary of visual words which will be invariant to changes in viewpoint and illumination.For this,

we used the SIFT descriptor[24],which is one of the most widely used descriptor to extract local image features.SIFT descriptors,based on histograms of local orientation,has some tolerance to illumination change.

The experiments was conducted to evaluate the accuracy of2-gram and3-gram representations compared to1-gram representation.The tests was performed in two different tasks: image retrieval and image classi?cation.

A.Image retrieval evaluation

We performed an image retrieval evaluation on four different image databases,two public databases and two medical ones:?Corel1000database1:consists of images of natural

scenes.It is composed of1000images divided into10 classes,100images in each class.Figure5a shows an image for each class of this database;

?Lung database2:consists of CT images of lung, composed of234images divided into6classes(39 images in each class),classi?ed according to the Lung ?ndings:Emphysema,Honeycombing,Interlobular Septal,Healthy,Consolidation and Ground-glass.Figure 5b shows an image for each class of this database;?Medical Image Exams database3:it contains2,200 medical images of X-ray and MRI,classi?ed according to body part and type of cut,being composed of11distinct classes with200images in each class.Figure5c shows images samples from this database.

?Texture database[25]:includes surfaces composed of materials such as wood,marble and fur under varying viewpoints,scales and illumination conditions.

This database consists of1,000images comprising40 samples of25different textures.Figure5d shows images samples from this database.

The comparative retrieval performance was evaluated using the mean of Average Precision(mAP).The Precision is the ratio of the number of relevant retrieved images to the total number of retrieved images.The Average Precision computes the average value of Precision at each ranking position where a relevant image is retrieved.The top value is1,which means that all relevant images were retrieved in the?rst positions of the ranking.The mAP determine the mean of Average Precision considering each image as query.The best value of mAP is1.

Table I presents the average mAP values for the evaluated databases.The2-gram representation presented the best results in three databases.In general,the2-gram boosted the precision in5%in these databases.Except for the Corel1000database, the3-gram presented the best result.In this database,the3-gram representation increased13%the precision compared with the traditional1-gram representation.

Considering each class individually,we can see where the results were more signi?cant.Table III presents the mAP 1available at:https://www.wendangku.net/doc/4316772984.html,/docs/related/

2provide by the Clinical Hospital with our university.

3provide by the Clinical Hospital with our

university.

(a)

(b)

(c)

(d)

Figure 5.Sample images from the:(a)Corel1000database,(b)Lung database,(c)Medical Image Exams database,(d)Texture database. results for each class of the Lung database and Tables II,IV and V summarize the results from representative classes of the other three databases,where the difference between the methods were considered signi?cant.Figures6a,6b,6c and 6d present the Precision values for each class of the evaluated databases.

For the class Emphysema in the Lung database and for the class Knee in the Medical Image Exams database,the2-gram representation had a gain up to17%in Precision.For the class Africa in the Corel1000database,the2-gram representation improve the precision in26%.Considering the Class1of the Texture database the3-gram representation achieved a gain of 44%in precision.

0.0.0.Class

P r e c

i s i o n

Corel1000database

0.35

0.4

Class P r e c i s i o n

Lung database

0.3

0.40.50.6Class

P r e c i s i o n

Medical Image Exams database

0.4

0.6

0.8

Class

P r e c i s i o n

Texture database

Figure 6.

Precision values for each class of the evaluated databases.

Table I

A VERAGE RESULTS FOR THE EVALUATED DATABASES .

Database Image representation 1-gram 2-gram 3-gram Corel10000.2990.3320.340Lung 0.3650.3850.366Medical Exams

0.3980.4210.412Texture

0.648

0.681

0.667

In general,when a class is complex,this means,when it has more visual details,the n -gram representation with a large value of n tends to have better results.However,the value of n affects the dictionary size,such that more visual phrases are being analyzed and this can affect the retrieval performance.Considering the overall performance for all the evaluated databases,the 2-gram and 3-gram achieved a gain in Precision compared to the traditional 1-gram representation.These results demonstrated that the high-level image feature proposed in this work was able to improve the retrieval system and making a CBIR closer to the users’expectation.

Table II

P RECISION RESULTS FOR SOME CLASSES OF THE C OREL 1000DATABASE

Class

Image Representation 1-gram 2-gram 3-gram 2Food 0.1710.1890.2193Horses 0.1750.2150.1745Flower 0.2310.2810.3096Africa 0.3050.3850.3557Elephant 0.3060.3560.3499Mountain 0.4890.4720.52910

Dinosaur

0.616

0.671

0.690

B.Image classi?cation evaluation

We performed an image classi?cation evaluation using the 15-scenes database [14].This database (Fig.7)is composed of 4485images categorized in ?fteen different scenes.Each category has 200to 400images,and average image size is 300x250pixels.The major sources of the pictures in the database include the COREL collection,personal photographs,and Google image search.This is one of the most complete

Table III

P RECISION RESULTS FOR THE L UNG DATABASE

Class

Image representation 1-gram 2-gram 3-gram 1Emphysema 0.3040.3470.3302Honeycombing 0.3330.3760.3303Interlobular Septal

0.3640.3730.3394Healthy 0.3710.3750.3695Consolidation 0.3910.3960.3986

Ground-glass

0.4250.446

0.429

Table IV

P RECISION RESULTS FOR SOME CLASSES OF THE M EDICAL I MAGE

E XAMS DATABASE

Class Image representation 1-gram 2-gram 3-gram 1Chest 0.2370.2590.2492Brain Axial

0.2380.2760.2704Hand 0.2570.2580.2525Foot

0.3110.2980.3196Brain Coronal

0.4050.4190.4007Breast 0.4870.5350.4808Knee 0.5270.6160.59110Abdomen 0.5600.6480.64011

Brain Sagittal

0.566

0.549

0.558

scene category database used in the literature thus far.

A multi-class classi?cation was done with a support vector machine (SVM)trained using the one-versus-all rule:a classi?er is learned to separate each class from the rest,and a test image is assigned to the label of the classi?er with the highest response.

Table VI shows the classi?cation rate for the experiments using 100images per class for training and the rest for testing (the same setup as [14]).The 2-gram representation presented the best classi?cation rate with 48%,followed by the 3-gram representation (43%).Compared with the 1-gram,the 2-gram representation had a gain of 11%in accuracy.This result demonstrate that the 2-gram representation encode more discriminative features than the 1-gram representation.

Figure 7.Example images from the 15-scenes category database.

Table V

P RECISION RESULTS FOR SOME CLASSES OF THE T EXTURE DATABASE .

Class Image representation 1-gram 2-gram 3-gram 1Class 10.3140.4140.4522Class 20.4270.4520.4274Class 40.4930.5300.5275Class 50.4990.5270.5586Class 60.5150.5140.4867Class 70.5460.5910.5349Class 90.5930.5820.56810Class 100.5990.6190.62211Class 110.6180.6660.68013Class 130.6550.7030.68115Class 150.6690.7130.69516Class 160.7040.7740.71518Class 180.7250.8160.84819Class 190.7450.7850.72022Class 220.7720.8390.77823Class 230.8230.8870.91124Class 240.9260.9600.96725

Class 25

0.9640.955

0.935

Table VI

A VERAGE CLASSIFICATION RATE FOR THE 15-SCENES DATABASE .

Representation

Average Classi?cation

1-gram 0.4282-gram 0.4763-gram

0.434

Analyzing each class individually,Table VII presents the classi?cation rates for each class of this database.For three classes,the 3-gram presented the best results.The 2-gram had a gain of 33%in accuracy for the class Kitchen and the 3-gram had a gain of 20%for the class MITforest compared with the traditional 1-gram approach.

In this evaluation task (image classi?cation),we can see the same behavior that the previous task (image retrieval).The 2-gram and 3-gram representations presented an overall performance better than 1-gram representation.These results indicate that the 2-gram and 3-gram add more information in the image feature description,being a valuable asset to improve the image analysis.

V.C ONCLUSION

In this paper,we have introduced a novel modeling approach for representing images.The new approach represents an image by taking into consideration the relationship between its visual words.The proposed method is based on the idea of Bag-of-Visual-Phrases,which has a higher level of semantic characterization compared to the traditional Bag-of-Visual-Words.The image is represented as a collection of visual

phrases,instead of considering the image as a set of isolated visual words.

Our proposed method uses a dictionary composed of visual phrases with a?xed number of words.Such representation is an analogy to the popular n-gram representation used for textual representation.

We have conducted experiments in?ve different databases. Three of the databases are public and employed as benchmarks in the image retrieval and classi?cation community.The others two are composed of medical images to demonstrate an area that can bene?t from the proposed technique.The results have shown that bigram and trigram dictionaries are suf?cient to boost the retrieval and classi?cation accuracy.

Our proposed novel modeling approach enriches the Bag-of-Visual-Words representation and the obtained results indicate that it can become a powerful and promising descriptor for image representation,and can contribute to the content-based image retrieval and image classi?cation?eld.

Table VII

C LASSIFICATION RESULTS FOR EACH CLASS OF THE15-SCENE DATABASE.

T HE HIGHEST RESULTS FOR EACH CLASS ARE SHOWN IN BOLD.

Class

Representation

1-gram2-gram3-gram

bedroom0.2500.3190.216

CALsuburb0.7870.9010.872

industrial0.1990.2460.235

kitchen0.2450.3270.227

livingroom0.2170.2330.211

MITcoast0.4730.4000.408

MITforest0.7810.8600.943

MIThighway0.4190.3940.263

MITinsidecity0.4520.4860.453

MITmountain0.4380.5290.453

MITopencountry0.3650.3840.297

MITstreet0.3280.4320.313

MITtallbuilding0.4300.4800.484

PARof?ce0.5220.5830.552

store0.5210.5670.591

A CKNOWLEDGEMENTS

The authors acknowledge the?nancial support of FAPESP, CNPq,CAPES and INCT INCod.

R EFERENCES

[1]M.S.Lew,N.Sebe,and J.P.Eakins,“Challenges of image and video

retrieval,”in International Conference on Image and Video Retrieval (CIVR),2002,pp.1–6.

[2]J.C.Caicedo,A.Cruz-Roa,and F.A.Gonzalez,“Histopathology image

classi?cation using bag of features and kernel functions,”in Conference on Arti?cial Intelligence in Medicine,ser.Lecture Notes in Computer Science,vol.5651,2009,pp.126–135.

[3]H.Jégou,M.Douze,and C.Schmid,“Improving bag-of-features for

large scale image search,”International Journal of Computer Vision, vol.87,no.3,pp.316–336,2010.

[4] D.Nister and H.Stewenius,“Scalable recognition with a vocabulary

tree,”in IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2006,pp.2161–2168.

[5]J.Philbin,O.Chum,M.Isard,J.Sivic,and A.Zisserman,“Object

retrieval with large vocabularies and fast spatial matching,”in IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2007.

[6]M.M.Rahman,S.K.Antani,and G.R.Thoma,“Biomedical cbir

using"bag of keypoints"in a modi?ed inverted index,”in International Symposium on Computer-Based Medical Systems,ser.CBMS’11,2011, pp.1–6.

[7]J.Sivic and A.Zisserman,“Video google:A text retrieval approach to

object matching in videos,”in International Conference on Computer Vision,vol.2,Oct.2003,pp.1470–1477.

[8]J.Wang,Y.Li,Y.Zhang,H.Xie,and C.Wang,“Boosted learning

of visual word weighting factors for bag-of-features based medical image retrieval,”in Image and Graphics(ICIG),2011Sixth International Conference on,2011,pp.1035–1040.

[9]L.Torresani,M.Szummer,and A.W.Fitzgibbon,“Learning query-

dependent pre?lters for scalable image retrieval,”in IEEE Conference on Computer Vision and Pattern Recognition,ser.CVPR’09,2009,pp.

2615–2622.

[10]J.Yuan,Y.Wu,and M.Yang,“Discovery of collocation patterns:from

visual words to visual phrases,”in IEEE Conference on Computer Vision and Pattern Recognition,ser.CVPR’07.IEEE Computer Society,2007, pp.18–23.

[11]Y.Zhang,Z.Jia,and T.Chen,“Image retrieval with geometry-preserving

visual phrases,”in IEEE Conference on Computer Vision and Pattern Recognition,ser.CVPR’11.IEEE,2011,pp.809–816.

[12]L.C.Zitnick,J.Sun,R.Szeliski,and S.Winder,“Object instance

recognition using triplets of feature symbols,”in Tech.Report,Microsoft Research,2007.

[13] C.Y.Suen,“n-gram statistics for natural language understanding and

text processing,”IEEE Transactions on Pattern Analysis and Machine Intelligence,vol.1,no.2,pp.164–172,1979.

[14]https://www.wendangku.net/doc/4316772984.html,zebnik, C.Schmid,and J.Ponce,“Beyond bags of features:

Spatial pyramid matching for recognizing natural scene categories,”

in Proceedings of the2006IEEE Computer Society Conference on Computer Vision and Pattern Recognition-Volume2,ser.CVPR’06.

Washington,DC,USA:IEEE Computer Society,2006,pp.2169–2178.

[15]S.Savarese,J.Winn,and A.Criminisi,“Discriminative object class

models of appearance and shape by correlatons,”in In IEEE Computer Vision and Pattern Recognition,2006,pp.2033–2040.

[16]Y.Cao,C.Wang,Z.Li,L.Zhang,and L.Zhang,“Spatial-bag-of-

features,”in CVPR’10,2010,pp.3352–3359.

[17]O.A.B.Penatti,E.Valle,and R.da Silva Torres,“Encoding spatial

arrangement of visual words,”in CIARP,2011,pp.240–247.

[18]Y.Jiang,J.Meng,and J.Yuan,“Grid-based local feature bundling for

ef?cient object search and localization,”in Image Processing(ICIP), 201118th IEEE International Conference on,2011,pp.113–116. [19]J.Sivic, B.Russell, A.Efros, A.Zisserman,and W.Freeman,

“Discovering objects and their location in images,”in Computer Vision, 2005.ICCV2005.Tenth IEEE International Conference on,vol.1,2005, pp.370–377V ol.1.

[20] D.Liu,G.Hua,P.Viola,and T.Chen,“Integrated feature selection

and higher-order spatial feature extraction for object categorization,”

in Computer Vision and Pattern Recognition,2008.CVPR2008.IEEE Conference on,2008,pp.1–8.

[21]N.Morioka and S.Satoh,“Building compact local pairwise codebook

with joint feature space clustering,”in ECCV2010,ser.Lecture Notes in Computer Science,K.Daniilidis,P.Maragos,and N.Paragios,Eds.

Springer Berlin Heidelberg,2010,vol.6311,pp.692–705.

[22]T.Li,T.Mei,I.-S.Kweon,and X.-S.Hua,“Contextual bag-of-words

for visual categorization,”Circuits and Systems for Video Technology, IEEE Transactions on,vol.21,no.4,pp.381–392,2011.

[23]P.Tirilly,V.Claveau,and P.Gros,“Language modeling for bag-of-visual

words image categorization,”in Proceedings of the2008international conference on Content-based image and video retrieval,ser.CIVR’08.

New York,NY,USA:ACM,2008,pp.249–258.

[24] D.G.Lowe,“Distinctive image features from scale-invariant keypoints,”

International Journal of Computer Vision,vol.60,no.2,pp.91–110, Nov.2004.

[25]https://www.wendangku.net/doc/4316772984.html,zebnik,C.Schmid,and J.Ponce,“A sparse texture representation

using local af?ne regions,”IEEE Transactions on Pattern Analysis and Machine Intelligence,vol.27,pp.1265–1278,2005.

财务会计专业实践教学的反思

财务会计专业实践教学的反思【摘要】财务会计是一类实践性很强的专业，实践是不可缺少的重要环节，针对目前我国高职院校财务会计专业实践教学存在的问题，提出了具体的改革建议，对促进实践教学质量的提高具有重要的意义。【关键词】会计专业；财务会计；实践教学；思考会计是经济管理应用学科的一个分支，由于对实践能力要求比较高，作为现代高职院校的毕业生必须具备较强的动手能力，实践性教学是必不可少的内容, 也是会计教学的重要环节。财务会计实践教学在许多高职院校普遍受到重视，对实施素质教育，培养高水平的会计实用人才发挥了较好的作用。但也有不少院校的财务会计实践教学仅仅是简单的动手操作，没有模拟真实的实训环境，造成了财务会计理论与实践严重脱节，从而影响了实践教学的质量。因此，加强贬务会计专业实践教学策略的研究，对促进实践教学质量的提高具有重要的意义。一、目前财务会计专业实践教学存在的问题（一）实践教学体系不完善

首先，实践课程标准不合理。其次，实训周时间不足，实践教学操作步骤缺失。实践性教学一般是指模拟实习和校外企业顶岗实习，其目的是理论用于实际，培养学生的实际动手能力。但是在现行的财务会计实践性教学中，财务会计专业教学上教与学的模式是：学生听课做笔记，教师用PPT 授课和板书来代替教学内容的进一步讨论和深究，职业岗位意识不强，往往忽视了出纳岗位、银行岗位、成本会计岗位、税务会计岗位和审计岗位等的培训，对于其工作职责、工作内容不明确，动手操作却相对较少。（二）专业教材更新太慢目前大多数院校会计专业所使用的专业教材质量比较差，存在新税收，新企业会计准则没有及时变更等的问题。大家知道，会计方面的法规、准则、规定并不是一成不变的，往往会根据当前的形势加以更新修改部分的条款规定，因此，教师在编写教材时要与时俱进，根据最新的税法新规（营改增等内容）该更新的内容要及时更新。但目前我国公开出版的各类会计教材中，大多数的教材缺乏一个科学的标准，没有体现会计专业教育的特色，教材编写不规范，重复混乱，漏洞百出，相互抄袭严而且信息过等等， IE 重，在结构、内容上雷同精品的教材极少。因此，专业教材

会计专业实践教学方案(修订版)

会计专业实践教学方案（修订版）会计学本科专业重点培养的是面向基层的应用型高等专门人才，本专业融理论、技能于一体，专业性、应用性较强，而学生缺乏业务实践。为锻炼培养学生的实际操作能力，系统掌握财务、会计的基本理论、基本知识、基本方法，增强学生毕业后的工作适应性，在教学中必须重视实践性教学环节。本专业实践性教学环节包括认知实习、课程实训和毕业实习三部分内容。具体内容和要求如下：第一部分认知实习认知实习是通过介入会计核算与会计管理某个层面的实践，增加学生对会计核算与会计管理工作的感性认识，培养和训练学生认识和掌握会计技能的重要教学环节。第二部分课程实训本次实践教学方案为了培养学生的动手能力，在原有实践计划的基础上在几门主要专业课程上增设了实践课程，这也是这次实践方案的重点以及创新之处。具体包括：一、会计学原理实践二、商业银行会计模拟三、会计电算化四、会计手工模拟第三部分毕业实习毕业实习是学生学完所有专业课程，掌握了会计学科基本理论和方法，并经过一系列实践环节训练的基础上，开展的面向社会的专业实践。

第一部分认知实习认知实习是培养、训练学生观察社会、认识社会以及提高学员分析和解决问题能力的重要教学环节。通过组织学生集中到某一独立核算的企业或事业单位进行会计岗位现场实习，熟悉会计循环中账务处理程序的各个环节，以强化学生对财务会计基本操作程序和技能的掌握。它要求学生运用所学知识和专业技能，通过对单位会计核算和会计岗位现场进行考察和分析，形成调查报告。（一）要求 1、凡已修本专业课程学分达到全部课程总学分60%以上的学生，可参加社会调查。认知实习为各专业学生必修环节，不得免修。调查选题应与会计学专业相关，具体内容由各试点单位确定，教学点组织实施。 2、学生应根据社会调查的选题列出调查提纲（包括调查对象和调查内容），作好采访笔记。 3、根据调查结果写出调查报告或工作流程。（二）时间安排认知实习可在第四学期末布置，暑假期间参加社会调查、完成调查报告写作和考核工作。（三）考核与成绩评定 1、考核内容（1）认知实习时的基本情况与表现；（2）认知实习的内容与文字表达。认知实习的内容应包括：调查对象一般情况、调查内容，调查结果，调查体会。字数要求不少于2000字。调查报告要求语言简练、准确；叙述清楚、明白；数据、资料可靠；结论有理、有据。 2、考核标准认知实习成绩分为合格与不合格两个等级。 3、成绩评定办法（1）由个人填写鉴定表，由被调查单位盖章。（2）调查报告原则上由指导教师评定。（3）调查报告初评不及格者，允许补写一次。第二部分课程实训一、会计学原理实践《会计学原理》作为会计专业基础课程，除了理论教学以外还需要补充一些实践教学，例如会计循环操作实践：（一）实践目的观察、练习工业企业主要经营过程的核算；了解企业财务核算的流程及每个环节中的具体要求、方法和使用的资料；掌握不同核算形式下的内涵、方法及具体要求；了解财务会计报告的编制基础和方法。（二）实践内容 1、根据一系列经济业务填制和审查原始凭证； 2、编制相应的记账凭证，注意专业记账凭证和通用记账凭证的运用； 3、通过各种不同的核算形式完成同一组经济业务的核算流程； 4、引导学生一起分析、总计各种核算形式下的差异、优缺点及适用范围

From Head to Toe 中英文

From Head to Toe 从头到脚跟我学 I am a penguin and I turn my head. Can you do it? 我是企鹅，我会转头。你能做到吗？ I can do it! 我能做到！ I am a giraffe and I bend my neck. Can you do it? 我是长颈鹿，我会弯脖子。你能做到吗？ I can do it! 我能做到！ I am a buffalo and I raise my shoulders. Can you do it? 我是野牛，我会抖肩膀。你能做到吗？ I can do it! 我能做到！ I am a monkey and I wave my arms. Can you do it? 我是猴子，我会挥动手臂。你能做到吗？ I can do it! 我能做到！ I am a seal and I clap my hands. Can you do it? 我是海豹，我会拍手。你能做到吗？ I can do it! 我能做到！ I am a gorilla and I thump my chest. Can you do it? 我是猩猩，我会捶胸。你能做到吗？ I can do it! 我能做到！ I am a cat and I arch my back. Can you do it? 我是猫，我会弓背。你能做到吗？ I can do it! 我能做到！ I am a crocodile and I wriggle my hips. Can you do it? 我是鳄鱼，我会扭屁股。你能做到吗？ I can do it! 我能做到！ I am a camel and I bend my knees. Can you do it? 我是骆驼，我会弯膝盖。你能做到吗？ I can do it! 我能做到！ I am a donkey and I kick my legs. Can you do it? 我是驴子，我会踢腿。你能做到吗？ I can do it! 我能做到！ I am an elephant and I stomp my foot. Can you do it? 我是大象，我会跺脚。你能做到吗？ I can do it! 我能做到！ I am I and I wiggle my toe. Can you do it? 我是我，我会动我的脚趾。你能做到吗？ I can do it! I can do it! 我能做到！我能做到！

深圳牛津版九年级下Unit 4 随堂读写训练(包含答案)

Unit 4 读写训练阅读理解 CFP Several months ago, 8-year-old Alex McKelvey set out to do some nice things for people she didn’t even know. For the past year, since her grandmother Linda died, the Washington girl and her family have performed hundreds of acts of kindness. “This is what her life was about, giving to others,” Alex said. She recently reached a goal that helped her family and hundreds of people who’ve received their special gifts. Alex was up before dawn on Sunday volunteering in a local town. That alone could have been her good deed for the day. But she was just getting started. By 8 am, the McKelveys were ready for breakfast at a restaurant where they would complete their next task. Before they paid their bill, they asked the server to swipe their card for a couple sitting a few tables away. But Alex wasn’t done yet. She then handed her server a 100-dollar tip. “Thank you, you guys are too sweet!” the server said, holding back tears. But these good deeds aren’t as random (随机的) as you might think. The y’re part of an important list. “Sixty-one different acts of kindness,” said Sarah McKelvey, Alex’s mom. And that was just for Sunday. The McKelveys have taken a lot of time over the past year to make other people’s days. “It’s been fun to come up with all these ideas, but it also kind of helps us too,” said Sarah. In all, the McKelvey family completed 600 acts of kindness in time for what would have been Linda’s 60th birthday on Sunday. 1. What special thing did the grandmother do? A. She helped other people. B. She performed for other people. C. She volunteered at a local hospital. D. She gave others special gifts. 2. Where did the McKelveys complete their next task? A. At a library. B. At a restaurant. C. At a local town. D. At a local school. 3. How many acts of kindness did the McKelvey family complete? A. 100 B. 60 C. 600 D. 8 4. What else do we know about Alex? A. She only does nice things for people she knows. B. Alex gave a server a 100-dollar tip. C. She will continue to do nice things for others. D. She will celebrate her birthday by helping others. 完形填空

应用型本科会计专业实践教学模式的创新

应用型本科会计专业实践教学模式的创新研究田凤萍（黑龙江商业职工大学，黑龙江哈尔滨 150018） [摘要]为体现应用型会计专业实践教学创新模式的特点，高校在实践教学过程中，应将“以教师为中心，以书本为依据，以教室为基地”的教学模式转变为“以学生为中心，以需求为动因，以能力培养、综合素质的提高为根本，以服务社会为宗旨、以岗位为基础”的教学模式。应用型高校应从会计本科专业培养目标的定位出发，从实践教学目标、内容、管理、评价及实践教学支撑保障体系等方面建立一套行之有效实践教学体系，从而培养出具有实践技能及市场需要的应用型人才。 [关键词]应用型；会计专业；实践教学模式；创新研究[中图分类号]F230 [文献标识码]B [收稿日期]2011-12-18 [作者简介]田凤萍(1963-)，黑龙江哈尔滨人，黑龙江商业大学副教授，注册会计师。研究方向：会计学。会计专业是与社会经济发展联系十分紧密的应用型学科。随着高等教育的大众化趋势、经济和科技的发展，社会管理类职业群和企业管理职业岗位群发生了变化，呈现出理论比重持续上升、技术含量不断增加、能力要求多元化的趋势，尤其是高新技术产业的职业岗位变动的频繁，对会计专业尤其是会计专业人才提出了新的要求。不仅要求理论基础扎实，还要技术水平过关；不仅要具备较强的实践操作能力，还要有创新能力、外语能力、信息化管理能力、团队合作能力等多种能力。因此，面对新形势，会计专业的教育如何培养适应社会需要的应用型人才是本科高校面临的一个重要问题。一、会计专业实践教学创新模式的特点应用型会计专业实践教学创新模式的特点为，一条主线、两个体系、三个基础、四个方面、五个层面、六个环节的特点。具体为： “一条主线”体现了以岗位对会计专业从业人员的能力和基本素质的要求为目标；为实现这一目标，必须合理设置“两个体系” ———理论教学体系和实践教学体系，在理论教学体系里设置了公共基础课、专业基础课、专业核心课、选修课四个平台，提高学生综合素质，提供必需、够用的理论知识，奠定可持续发展的知识基础；“三大基础”在实践教学体系里设置了课程设计、实训、实习、实践协作综合实训、顶岗实习等子模块，为了体现会计专业特色，根据会计的发展现状和发展前景，必须打好“三大基础”：一是人文素质基础，它包括英语水平、财经写作及会计从业人员所必须的诚信基础和其他基本素质。二是会计业务处理基础，包括会计账务处理、会计报表的编制、经济活动的分析、资金的筹集以及预测和决策分析。三是计算机应用基础，它包括计算机操作与维护、财务软件的使用与维护等。 “四个方面”会计专业的实践教学应四个方面逐步展开：首先是科学构建实验实践教学内容体系，依次是实践教学目标、实践教学环节和实践教学结果的评价；针对目前本科会计学专业人才实践能力特别是综合实践能力、创新创业能力不强的实际情况，应将实验实践教学的功能定位，构建了一套涵盖课程单项性实验、课程综合性实验、专业综合性实验、跨专业综合性实验、创新创业实践 “五个层面”；实践教学包括实训、实验、社会调研、专业实习、综合实习(毕业实习)、毕业论文(设计)“六个环节”。二、积极探索应用型会计实践教学模式的创新之路 (一)以学生为中心会计实践教学是以先进的现代教育理论为指导，充分发挥学生在实践过程中的主动性、积极性、创造性，使学生成为实践活动的主体和知识意义的主动建构者。而不是外部刺激的被动接受者和知识灌输的对象；教师则成为课堂教学的组织者、指导者、学生建构意义的帮助者、促进者，而不是知识的灌输者和课堂的主宰者。 (二)以需求为动因其是在会计实践教学中及时了解社会和用人单位对会计人员的需求状况，预测社会经济发展对会计人员的知识、技能等方面提出的新要求，在此基础上确定会计实第2012年第2期（总第391期）商业经济 SHANGYE JINGJI No.2，2012Total No.391 [文章编号]1009-6043 （2012）02-0110-02110--

keys to vocabulary exercises unit 5

1. London looks like a middle aged gentleman. He's knowledgeable and ________ dressed. A. smartly B. lavishly铺张地；奢侈地 C. diplomatically D. agreeably Correct answer: A 2. If only some handsome boy invited me to be his ________ at the ball! A. competitor B. diplomat C. partner D. rival 对手 Correct answer: C 3. Friends said last night that his legal ability and ________ would add a new dimension to the negotiations. A. ambition B. competition C. loyalty D. intellect Correct answer: D 4. This disease is a ________, degenerative brain disorder, which gradually destroys the ability to reason, remember, imagine and learn. A. progressive B. threatening C. sophisticated D. impatient Correct answer: A 5. She won an Oscar ________ for best supporting actress. A. internship实习岗位;实习期 B. partnership C. competition D. nomination Correct answer: D 6. McCain already has about half of the delegates he needs to win the party ________. A. proposition命题;主张;建议 B. successor C. loyalty D. nomination Correct answer: D

会计专业实践教学创新研究报告

会会会会会会会会会会会会报告 ①课题的核心概念及其界定教学设计是根据课程标准的要求和教学对象的特点，将教学诸要素有序安排，确定合适的教学方案的设想和计划。一般包括教学目标、教学重难点、教学方法、教学步骤与时间分配等环节。是从工作实际出发，把教学原理转化为教学材料和教学活动的计划，是实现教学目标的计划性和决策性活动，以解决教什么的问题，怎样教的问题。以计划和布局安排的形式，对怎样才能达到教学目标进行创造性的决策，使教学效果最优化。本课题是基于我院会计专业，设计开发适合于会计专业教学使用的教学方案，从企业财务岗位工作任务角度进行探索，通过深入多个企业调研，与企业岗位对接，与企业专家实践座谈，明确会计从业人员需要胜任的技能点，根据会计助理、出纳、会计、财务主管等实际会计岗位专业技能要求，提炼为七个典型工作任务：“业务单据的分类与整理”、“货币资金往来业务核算”、“日常经济业务的处理”、“成本费用的核算”、“会计报表的编制与分析”、“税费的计算与申报”、“企业财务分析与决策”，将提炼的典型工作任务，结合学院实际，作为学院的专业课程进行转化。《会计日常经济业务处理》是其中第三个典型工作任务，技能任务涵盖资金收付、往来款项、存货、固定资产以及进销存的业务核算，是从事会计工作必备技能。因此以该典型工作任务为支撑点，组成教研团队，进一步提炼代表性工作任务，分配学习活动，设计教学方案，采用一体化教学实践，以学生为主体、能力为本位，以培养学生实际工作能力为目标，将理论融合实践活动，将企业实际任务的解决来促进学生对于专业知识、技能的掌握，真正做到“工学一体，学以致用”，实现“学生有兴趣、教学有效果、任务能解决”。 ②研究意义该课题的开发和研究意义表现在以下几个方面：课题开发过程中，通过企业调研、典型工作任务和代表性工作任务的提炼，收集某一行业企业发展的基础数据，并对该行业领域职业工作和职业教育状况进行观察分析，进一步深化认识，思考职业领域对会计专业人才的要求，完善培养目标和课程体系。教学设计的编写、完善过程，是专业教师以现代教学理论为基础，依据教学对象的特点和教师自己的教学观念、经验、风格，运用系统的观点与方法，分析教学中的问题和需要，确定教学目标，建立解决问题的步骤，合理组合和安排各种教学要素，为优化教学效果而制 39

会计专业实践教学体系探索

会计专业实践教学体系探索作者：高秀兰李雅丽来源：《财会通讯》2011年第04期一、会计专业大学生实践能力培养问题研究现状伴随着大学扩招的逐步深入，社会人才的供求关系出现了极大的变化，大学生就业市场逐步由卖方市场转向买方市场，大学生也越来越成为了就业的弱势群体，而作为招聘方的企业也在不断地抱怨新进大学生缺乏应有的实践能力和动手能力。因此，如何使更多的普通高校毕业生适应社会需求，提高自己的实践能力，在就业市场上找到自己的合适位置，是当代大学生面临的严峻考验，也是普通高等院校应用型人才培养需进一步研究的重要课题。基于这一现实，学术界对大学生实践能力相关问题进行了许多有益的探索。本文整理了与经管类会计学专业大学生实践能力相关的文献，对大学生实践能力的概念和内涵有一个清晰的了解；同时调研了唐山市多家招聘单位的招聘要求，从招聘单位的视角分析大学生应具备的实践能力；并结合分析经济管理类会计专业毕业生所认为的大学生应具备的实践能力，考察了招聘方要求和学生观点之间的差别，以期对经济管理类会计学本科专业大学生的人才培养方案等制定产生积极影响。（一）大学生实践能力的构成大学生实践能力主要由一般实践能力和专业实践能力构成。其中，一般实践能力指所有大学生在学习、生活及未来工作、人际关系处理等方面应具备的基本实践能力，主要包括学习能力、表达能力、沟通协调能力、心理健康能力、团队协作能力、身体健康能力、组织管理能力、吃苦耐劳能力、计算机应用能力、外在表现能力、内在品行能力、一般经验能力、诚实勤奋精神、敏捷思维能力、勇于开拓精神、创新能力、应变能力、逻辑能力等方面；专业实践能力则指专业大学生在学习、生活及未来工作、人际关系处理等方面应具备的专业性实践能力。（二）经济管理类大学生实践能力的构成经济管理类大学生除了具备所有大学生均具备的一些基本特点外，还体现在他们接受了经济管理类的专业学习和初步科研训练，同时还表现在他们未来所从事的工作将基本限于具有经济管理行业背景的企业或者其他企业的经济管理类岗位。经济管理类大学生的实践能力将必须首先满足大学生实践能力的一般实践能力的基本要求，即具备学习能力、表达能力、沟通协调能力、心理健康能力、团队协作能力、身体健康能力、组织管理能力、吃苦耐劳能力、计算机应用能力、外在表现能力、内在品行能力、一般经验能力、诚实勤奋精神、敏捷思维能力、勇于开拓精神、创新能力、应变能力、逻辑能力等方面的能力。同时，经济管理类大学生还必须具备包含专业经验能力、专业态度能力、专业工作能力、专业知识能力、专业道德能力和专业创新能力等在内的专业实践能力。所谓专业经验能力是指大学生在日常学习和生活中积累起来的与未来工作相关的专业经验；所谓专业态度能力是指大学生在日常学习和生活中培养形成的与未来工作相关的专业态度，主要包括工作责任心、勤恳敬业精神、细心认真精神、工作积极性和主动性、工作态度及事业心等方面；所谓专业知识能力是指大学生在日常学习和生活中积累起来的与未来工作相关的专业知识，从专业知

会计专业实践教学中的不足及措施

会计专业实践教学中的不足及措施实践教学是提高学生综合素质与创新能力的重要手段。但是，我国高校会计实践教学存在诸多问题，必须充分认识会计专业实践教学的重要性，采取各种措施，不断提高会计实践教学水平。一、会计专业实践教学的必要性加强会计专业实践教学是现代经济社会发展对会计人才培养提出的客观要求，是满足国际发展的需要，是会计专业人才培养目标的需要 (一)加强实践教学是建设创新型国家对人才培养提出的客观要求。建设创新型国家，科技是关键，人才是核心，教育是基础。面对建设创新型国家的目标，高等院校在会计教育方面必须全面推进素质教育，这就意味着高校培养出来的人才不仅要具有较高的理论素养和扎实的基本功，更要具有远见卓识和较强的创新能力。这种创新能力具体体现在高校会计专业的教学过程中，就是要对学生进行以专业基础知识教育为基础的综合素质训练，培养学生发现问题、分析问题、解决问题的能力。 (二)加强实践教学是与国际会计教育接轨的需要。作为培养会计创新人才重要基地的高等院校应当充分认识到加强实践教学对学生创新能力的培养具有重要意义。参与世界竞争、加入全球化的洪流已经成为普通中国人都认可的共识。在初步参与国际交往和实际竞争之后，发现我国高校以往建立起来的人才培养模式和人才标准在许多方面还难以做到“与国际接轨”。我国的会计人才要走向世界，会计教育要与国际接轨，高校会计专业必须加强实践教学，以提高学生的综合素质与创新能力。

(三)加强实践教学是会计专业人才培养目标与现实经济发展相结合的需要。会计专业属于应用性较强的学科，其培养的学生，除少数进入硕士研究生行列外，大部分将进入企业、事业单位和行政机关从事经济方面的种种工作。也就是说，高校会计专业所培养的学生主要是应用型人才，而非理论研究型人才，这就要求高校会计专业培养的学生不仅要具备较扎实的理论基础，更要具备相应的专业技能和素养，具有创新精神、实践能力和创新能力。加强实践教学是实现这一目标的非常重要的途径。二、目前会计实践教学中存在的问题改革开放以来，特别是高校扩招以后，我国会计专业发展迅速，但在高速发展的同时，各种问题也日渐显现。 (一)认识不足，重视不够。受传统教育思想的影响人，们往往认为高校教学应以理论教学为主，实践教学为辅；理论教学体现了学术性，实践教学是理论教学的一个补充。在会计专业的人才培养过程中，表现为重理论轻实践、重知识轻能力、重课堂轻课外、重定性轻定量等倾向。实践教学在会计专业没有得到应有的重视。因此，不可避免地造成学生在理论与实践上的脱节，从而导致学生动手能力不强，会计专业的毕业生面临越来越大的就业压力。在专业教学中，学校首先考虑的是如何安排理论课的教学任务，实践教学往往被放在次要位臵上。 (二)教师会计实践水平不高。我国高校从事会计教育的教师队伍中拥有～定企业实践和社会实践经验的老师不多，有丰富实践经验的会计专业人才到高校任教的也不多。绝大部分是从学校到讲台，从学生到教师，没有参加过相应的实践锻炼，缺乏丰富的社会会计实践经历。在讲授专业课时不能很好地引用实践案例阐述基本理论和原理，

一首好听的儿歌From head to toe的歌词

From head to toe Let’s learn the song I am a penguin and I turn my head Can you do it? I can do it I am a giraffe and I back my neck Can you do it? I can do it I am a buffalo and I raise my shoulders Can you do it? I can do it I am a monkey and I wave my arms Can you do it? I can do it I am a seel and I clap my hands Can you do it? I can do it I am a gorilla and I thumb my cheast Can you do it? I can do it I am a cat and I arch my back Can you do it? I can do it I am a crocodile and I rick my hips Can you do it? I can do it I am a camel and I bent my knees Can you do it? I can do it I am a donkey and I kick my legs Can you do it? I can do it I am a elephant and I stump my foot Can you do it? I can do it I am I and I raise my toe Can you do it? I can do it I can do it!

会计专业实践教学中的不足及措施

会计专业实践教学中的不足及措施提要本文就会计专业实践教学存在的问题,分析其原因,并提出建议。关键词:会计;实践教学;专业人才中图分类号:G64文献标识码:A 实践教学是提高学生综合素质与创新能力的重要手段。但是,我国高校会计实践教学存在诸多问题,必须充分认识会计专业实践教学的重要性,采取各种措施,不断提高会计实践教学水平。一、会计专业实践教学的必要性加强会计专业实践教学是现代经济社会发展对会计人才培养提出的客观要求,是满足国际发展的需要,是会计专业人才培养目标的需要。 (一)加强实践教学是建设创新型国家对人才培养提出的客观要求。建设创新型国家,科技是关键,人才是核心,教育是基础。面对建设创新型国家的目标,高等院校在会计教育方面必须全面推进素质教育,这就意味着高校培养出来的人才不仅要具有较高的理论素养和扎实的基本功,更要具有远见卓识和较强的创新能力。这种创新能力具体体现在高校会计专业的教学过程中,就是要对学生进行以专业基础知识教育为基础的综合素质训练,培养学生发现问题、分析问题、解决问题的能力。 (二)加强实践教学是与国际会计教育接轨的需要。作为培养会计创新人才重要基地的高等院校应当充分认识到加强实践教学对学生创新能力的培养具有重要意义。参与世界竞争、加入全球化的洪流已经成为普通中国人都认可的共识。在初步参与国际交往和实际竞争之后,发现我国高校以往建立起来的人才培养模式和人才标准在许多方面还难以做到“与国际接轨”。我国的会计人才要走向世界,会计教育要与国际接轨,高校会计专业必须加强实践教学,以提高学生的综合素质与创新能力。 (三)加强实践教学是会计专业人才培养目标与现实经济发展相结合的需要。会计专业属于应用性较强的学科,其培养的学生,除少数进入硕士研究生行列外,大部分将进入企业、事业单位和行政机关从事经济方面的种种工作。也就是说,高校会计专业所培养的学生主要是应用型人才,而非理论研究型人才,这就要求高

会计实践教学问题的完善策略探究【论文】

会计实践教学问题的完善策略探究摘要：在我国社会主义市场经济迅猛发展的背景下，企业基于自身市场竞争力提升的目的，急需各类经济学专业人才进入到企业当中，借助其专业化的经济学知识来帮助企业分析当前经济局势、调整企业经济发展规划，进而实现新经济环境下企业创新改革发展目标的实现。这就对经济学人才培养重要基地的高校会计专业提出了更高的要求，不仅要求会计专业在教学上能够充分地体现职业人才培养目标，同时更要求会计专业必须积极建立以市场为导向的人才培养体系，以强化人才培养的实用性。为此，在此背景下，本文以高校会计专业教学为研究对象，通过对会计实践教学过程中存在问题的分析，探究了当前环境下高校会计实践教学的完善策略。希望借助以下研究和探讨对进一步促进会计专业教学创新改革的实现具有一定的积极意义。关键词：会计；实践教学；问题；策略实践教学与理论教学共同组成会计教学完整体系，在会计教学中都占据着重要的位置，并各自发挥着不同的教学功能。会计理论教学主要是对学生进行会计相关概念、方法等的知识性传授，是学生开展会计工作的重要基础；会计实践

教学则主要是对学生进行会计实际操作、市场环境分析等方面的技能、经验的传授，是学生会计工作质量、效率的重要保障。相对而言，受以往传统教学观念及实践教学设施、体系不完善等情况的限制，使得会计实践在以往的教学中一直未受到应有的重视，从而导致目前高校会计实践教学问题突出，严重影响了高校会计专业教学的整体发展，同时也不利于会计人才岗位适应能力的提升。因此，在当今社会主义市场经济迅猛发展以及会计新课改的共同要求下，我们对会计实践教学展开研究与探讨，无论是对于高校自身发展还是对于促进社会经济变革都将有着极为重要的现实意义。 1会计实践教学中存在的问题 1.1学生职业能力比较缺乏基于企业自身多方面发展的需求，很多企业对于会计专业人才的职业技能提出了更高的要求，但是本文通过对目前会计实践教学的现状调查得知，学生职业能力普遍比较缺乏，无法满足企业在经济发展方面的各种需求。主要体现在以下几个方面：第一，会计实践经验匮乏。对于很多会计专业学生来讲，在开展实践教学之前，其仅仅接受过专业的理论知识、技能教学，而对于实践教学的参与则机会寥寥。因

From Bag-of-Visual-Words to Bag-of-Visual-Phrases Using__ n-Grams

财务会计专业实践教学的反思

会计专业实践教学方案(修订版)

From Head to Toe 中英文

深圳牛津版九年级下Unit 4 随堂读写训练(包含答案)

应用型本科会计专业实践教学模式的创新

keys to vocabulary exercises unit 5

会计专业实践教学创新研究报告

会计专业实践教学体系探索

最新新视野大学英语第二版第二册课文翻译Unit-3-Section-B

最新会计专业实践教学大纲

会计专业实践教学中的不足及措施

一首好听的儿歌From head to toe的歌词

会计专业实践教学中的不足及措施

会计实践教学问题的完善策略探究【论文】